ATLAScan Help

Optech ATLAScan Help
Preliminary draft for ATLAScan 1.0.0.81.

Copyright © 2017 by Teledyne Optech Incorporated. All Rights Reserved.
Table of contents
Introduction ................................................................................................................. 7
Contents .................................................................................................................. 7
System Requirements................................................................................................ 8
Installation ............................................................................................................... 9
License Manager ......................................................................................................13
Disclaimer ...............................................................................................................15
Launcher .....................................................................................................................16
Recent Projects Window ...........................................................................................17
Project Info Window.................................................................................................18
Control ........................................................................................................................19
Control Window, Simple ...........................................................................................20
Control Window, Advanced .......................................................................................21
Manage Surveys ......................................................................................................22
File .............................................................................................................................23
Create New Project Window......................................................................................24
Project Management ................................................................................................25
Import ....................................................................................................................26
Import Formats ...................................................................................................27
CAD formats....................................................................................................29
IFC .............................................................................................................30
Import ATLAScan Project ......................................................................................32
Load item............................................................................................................33
Export .....................................................................................................................34
Export Model as ...................................................................................................35
Export polyline.....................................................................................................39
Export cross section as .........................................................................................39
Export UCS to AutoCAD ........................................................................................40
Export annotations to CSV ....................................................................................41
Export pose .........................................................................................................42
Export to Autodesk ReCap 360 ..............................................................................43
Navigation ...................................................................................................................44
Predefined Views .....................................................................................................45
Slice View (Process Environment) ..........................................................................46
Horizontal Section View (Process Environment) ......................................................48
Navigation Modes ....................................................................................................48
Pivot mode ..........................................................................................................49
Walk mode ..........................................................................................................50
Bubble mode .......................................................................................................51
Zoom ..................................................................................................................52
Move to viewport .................................................................................................53

2 / 275 Copyright © 2017 by Teledyne Optech Incorporated. All Rights Reserved.
Recent camera positions.......................................................................................54

Screen Settings........................................................................................................55
Display and navigation Options .............................................................................56
View Parameters ......................................................................................................59
Creation of User Coordinate System ..........................................................................60
Outputs .......................................................................................................................62
Measures & Notes ....................................................................................................63
Elevations & Plans....................................................................................................64
Create Orthophoto ...............................................................................................65
Create Camera ....................................................................................................67
Place here survey point ........................................................................................68
Virtual scan .........................................................................................................69
Exporting orthophotos to AutoCAD ....................................................................71
Cross Sections .........................................................................................................72
Cross Sections .....................................................................................................73
Cross sections from plane collection ......................................................................75
Area & Volume ........................................................................................................76
Area ...................................................................................................................77
Volume ...............................................................................................................78
Volume PDF report ..........................................................................................80
Compute Cut & Fill Volume ...................................................................................81
Cut and fill by altitude report ............................................................................83
Inspection ...............................................................................................................84
Geometric Shapes Tools ...........................................................................................87
Create/edit plane .................................................................................................88
Create/edit segment ............................................................................................92
Create/edit circle .................................................................................................94
Create/edit polyline ..............................................................................................96
Create/edit sphere ...............................................................................................98
Create/edit cylinder ............................................................................................ 100
Create/edit cone ................................................................................................ 102
Distance between shapes ................................................................................... 104
Intersect geometric shapes ................................................................................. 105
Edit points list.................................................................................................... 106
Polyline from point list ........................................................................................ 107
Flythrough from point list ................................................................................... 108
Tunnel survey ........................................................................................................ 109
Video Record ......................................................................................................... 110
Flythrough Window ............................................................................................ 111
Movie Dialog ..................................................................................................... 112
Drawing tools ........................................................................................................ 113
Save snapshot ....................................................................................................... 114

Copyright © 2017 by Teledyne Optech Incorporated. All Rights Reserved. 3 / 275
Manage units of measures ...................................................................................... 115

Tools ........................................................................................................................ 118
Process Environment .............................................................................................. 119
File (Process Environment) ................................................................................. 121
Scan Processing Wizard .................................................................................. 122
Navigation (Process Environment) ....................................................................... 123
Tools (Process Environment)............................................................................... 124
Point Filtering & Clustering ............................................................................. 125
Pose & Registration........................................................................................ 126
Automatic Target-free Pre-registration......................................................... 127
Sketch Window ..................................................................................... 130
Edit Registration Constraints Window ...................................................... 132
Automatic Target-based Pre-registration ...................................................... 133
Target Editor ......................................................................................... 134
Targets Registration Report .................................................................... 135
Manual Pre-registration among models ........................................................ 136
Manual Pre-registration among grid point clouds .......................................... 138
Cloud to Cloud Registration ........................................................................ 140
Bundle Adjustment .................................................................................... 144
Scan alignment per groups ......................................................................... 149
Point cloud Georeferencing ......................................................................... 151
Manage Control Points ............................................................................... 153
Pose Transform ......................................................................................... 154
Manage circular targets .............................................................................. 155
Determine the cloud's vertical direction ....................................................... 156
Manual positioning ..................................................................................... 157
Select ........................................................................................................... 158
Measures & notes .......................................................................................... 159
Settings ........................................................................................................ 160
Wizard .......................................................................................................... 161
Processing Settings Dialog .......................................................................... 162
Parse ................................................................................................................ 163
Pose & Registration ................................................................................................ 164
Define project reference system .......................................................................... 165
Manual positioning ............................................................................................. 166
Create/edit UCS ................................................................................................. 167
Georeferencing from point list ............................................................................. 169
Registration report dialog ............................................................................... 170
Pose transform .................................................................................................. 171
Pose Dialog ....................................................................................................... 173
Selection & Fitting Tools ......................................................................................... 176

Select ............................................................................................................... 177

Point selection with polyline ............................................................................ 179
Sample to new cloud ...................................................................................... 180
Fitting ............................................................................................................... 181
Point Filtering & Clustering ..................................................................................... 182
Pre-processing Grids .......................................................................................... 183
Restore raw data ............................................................................................... 186
Restore deleted points ....................................................................................... 187
Edit 2D ............................................................................................................. 188
Point clustering .................................................................................................. 190
Make single cloud .............................................................................................. 192
Fill holes ........................................................................................................... 193
Hide black points ............................................................................................... 194
Remove duplicate points .................................................................................... 195
Resample .......................................................................................................... 196
Simplify points ................................................................................................... 197
Extract edges .................................................................................................... 198
Meshing ................................................................................................................ 199
Multiresolution Meshing ...................................................................................... 200
Meshing from predefined view ............................................................................ 204
3D Meshing ....................................................................................................... 206
Topographic meshing ......................................................................................... 207
Topographic meshing parameters ................................................................... 209
Mesh editing...................................................................................................... 211
Mesh Editor ................................................................................................... 212
Mesh selection from current view point............................................................ 216
Mesh selection with 3D polyline ...................................................................... 217
Make single mesh .......................................................................................... 218
Photo & Color ........................................................................................................ 219
Histogram ......................................................................................................... 221
Colors Mapping .................................................................................................. 222
Color with altitude.............................................................................................. 227
Inclination from plane ........................................................................................ 228
Color clouds with photos .................................................................................... 229
Create projector ................................................................................................ 230
Edit (projector) image .................................................................................... 231
Camera calibration on a grid ............................................................................... 232
Camera calibration using 3D points ..................................................................... 234
Calibrate camera intrinsics .................................................................................. 235
Panorama creator .............................................................................................. 237
Import GeoTIFF ................................................................................................. 238
Create texture map on a mesh ............................................................................ 239

Windows ................................................................................................................... 240

Project Window ..................................................................................................... 241
Recently used items dialog ................................................................................. 243
Property Window ................................................................................................... 244
Information Window .............................................................................................. 245
Readout Window ................................................................................................... 246
Log Window .......................................................................................................... 247
GUI options ........................................................................................................... 248
Custom Toolbar Editor ............................................................................................ 249
Workspace ................................................................................................................ 250
Help .......................................................................................................................... 251
Online help ............................................................................................................ 252
About plugins ........................................................................................................ 253
About ATLAScan .................................................................................................... 254
General Concepts ....................................................................................................... 255
Annotations ........................................................................................................... 256
Cameras ............................................................................................................... 257
Flythroughs ........................................................................................................... 259
Groups .................................................................................................................. 261
Planes ................................................................................................................... 263
Point Clouds .......................................................................................................... 264
Polylines................................................................................................................ 267
Projectors.............................................................................................................. 268
Triangle Meshes..................................................................................................... 269
User Coordinate System (UCS) ................................................................................ 272
Create UCS from this pose .................................................................................. 274

ATLAScan Manual
Optech ATLAScan is a software tool for extracting raw data from the Polaris Terrestrial Laser Scanner, preparing
intermediate representations, and extracting deliverable information. It can also optionally operate the Polaris directly.
ATLAScan's main advantage is its comprehensive toolkit and features that enable easy extraction of quality results in
many fields: mining, surveying, construction, architecture, cultural heritage, forensics, BIM, tunneling, etc.
1. First, ATLAScan the survey is collected using the Polaris. The operator normally executes the survey using the
on-board UI (OBUI) on the Polaris' touchscreen. However, if they have a laptop computer running ATLAScan
with them, they can also execute the survey using ATLAScan itself. This provides greater control over the survey
parameters than the OBUI.
2. ATLAScan imports the survey data onto the processing computer.
3. The processor uses ATLAScan's Process environment, a robust wizard, to easily import, pre-process,
and automatically register any amount of data with state-of-the-art speed and precision.
4. After that, the operator processes the data to extract basic results, including basic measures and
notes, elevations and plans, cross sections, areas and volumes, geometrical check and change detection, tunnel
analyses, flythrough videos, and so on.
5. Furthermore, ATLAScan provides all the features needed to deliver more elaborate results, such as tools
for model positioning, point cloud editing, fitting geometric primitives, model filtering and
clustering, meshing, mesh editing, coloring with calibrated photos, etc.
Please click on one of the following sub-topics:

• Contents
• System Requirements
• Installation
• License Manager
• Languages
• Disclaimers
• End User License Agreement

Contents
The ATLAScan online help is organized in four parts:
i. An Introduction on installing and licensing the software

ii. The main features of the User Interface
iii. An explanation of the functionalities of ATLAScan, divided according to the menu items, including the following main
chapters:
• File
• Navigation
• Outputs
• Tools
• Windows
• Workspaces
• Help
Note: Some commands are not available in all the workspaces, depending on which ATLAScan modules you have
licensed.

System requirements
System requirements — Minimum

• OS: Windows (XP SP2, Vista, 7, 8, 10)
• Version: 32 bit
• Graphic card: NVIDIA GeForce with 512 MB of RAM
• RAM: 4 GB
System requirements — Recommended

• OS: Windows (XP SP2, Vista, 7, 8, 10)
• Version: 64 bit
• CPU: Multi-core processor (at least 8 cores)
• Graphic card:
o NVIDIA GeForce GTX with 2 GB of RAM (for heavy usage of points)
o NVIDIA Quadro (for heavy usage of meshes and textures)
• RAM: 16 GB
ATLAScan compiled for 32-bit processors can only address 4GB of RAM memory. For medium/large size projects, we
strongly recommend the 64-bit version. The graphics card must be at least OpenGL 1.3 compliant with
ARB_transpose_matrix, ARB_multitexture, ARB_pbuffer extensions. Other extensions are automatically used if hardware
support is found, resulting in better performance and increased functionality.

Installation
It is strongly recommended that you deactivate any antivirus and antispyware software and close any existing Optech
ATLAScan application before starting the installation.
1. Run the installation application.

2. In the ATLAScan Setup window, click Next.
3. Accept the License Agreement terms and click Next.

4. Select the components you want to install and click Next. Note that the hardware drivers must be always
installed.
5. Choose Install Location (the default path is C:\Program Files\Teledyne Optech\ATLAScan (64 bit)) and click Next.

6. Choose a Start Menu Folder and click Next.
See also Help > About ATLAScan to find more information about the license, including the expiration date.

Setting up your environment

When you use ATLAScan for the first time, you must set certain options:
• Set your licensed Workspace (Works, Works Pro, Photo, etc.)
• If your PC has more than one graphic card (e.g. both an integrated and a dedicated graphic card), combine the
software with the better graphic card by using its control panel.

License Manager
Activation and updating software are performed in different ways, depending on the software license. There are
different procedures for:
• Activating a Software License
• Updating a Software License
Activating the Software License

You can activate the software license on a particular computer by generating a C2V file, sending it to Optech, and
applying the V2C file that Optech returns to you.
1. Open ATLAScan on the computer that you wish to license.

2. Select Help > License Manager.
3. Click Create .c2v license file in the License Manager window.

4. In the Collect License Information dialog, click Install new SL key.
5. Enter a name for the C2V file and a place to save it.

6. Sent the C2V file to Optech. Optech will send you a V2C license activation file.
7. In ATLAScan's License Manager window, click Apply .v2c license activation file.
8. Select the V2C file from Optech and click Open.
ATLAScan is now licensed on the computer.
Updating the Software License

You can also update the software license on a computer following these steps, which are very similar to the initial
activation steps:
1. Open ATLAScan on the computer that you wish to license.

2. Select Help > License Manager.
3. Click Create .c2v license file in the License Manager window.
4. In the Collect License Information dialog, click Update existing SL/HL key.
5. In the Locally Accessible Keys dialog, select your license from the drop-down menu and click OK.
6. Enter a name for the C2V file and a place to save it.
7. Sent the C2V file to Optech. Optech will send you a V2C license activation file.
8. In ATLAScan's License Manager window, click Apply .v2c license activation file.
9. Select the V2C file from Optech and click Open.

Disclaimers
Copyright © 2017 Teledyne Optech Incorporated

All rights reserved. This item and the information contained herein are the property of Teledyne Optech Incorporated.
No part of this document may be reproduced, transmitted, transcribed, stored in a retrieval system, or translated into
any language or computer language in any form or by any means otherwise, without the express written permission of
Teledyne Optech Incorporated, 300 Interchange Way, Vaughan, Ontario, Canada L4K 5Z8.
Disclaimer and Limitation of Liability

Teledyne Optech Incorporated assumes no responsibility for any damage, loss or other claim incurred by an end user or
any third party arising or resulting from the use of this software.
Trademark
Products and brand names may be trademarks or registered trademarks of their respective owners.
See also the End User License Agreement.

Project Launcher
The project launcher is the first screen that appears when you open ATLAScan. It enables you to create a new project,
navigate to an existing project, or select a recent project.
• If you open ATLAScan by double-clicking on a project file, ATLAScan will open that project in the Project Info
Window.
• If you open ATLAScan by double-clicking on its icon, ATLAScan opens the Recent Projects Window.

Recent Projects Window

The Recent Projects window displays several of your most recent or favorite ATLAScan projects. Clicking the preview
image of any of the projects opens that project in the Project Info window.
Recent Projects tools

Each project displays the following information and tools:
• You can click the lock icon at the top left to pin the project, which makes it always appear at the top of the list
of projects.
• The central panel shows a preview image of the project. The image used can be customized in the Project Info
window.
• You can removes the project from the recent projects list by clicking the red x at the top right. This does not
delete the project itself.
• The text below the preview image shows the project's name, the last time it was edited, and a user-created
description, defined in the Project Info window.

Project Info Window

The Project Info window shows details of the selected ATLAScan project.
Project Info tools

The window displays the following information and tools:
• Top bar: The top toolbar lets you create a new project, open an existing project, or open the current project
in the Control, Process or Extract environments.
• Recent Projects: Closes this project and opens the Recent Projects window.
• Project preview: An image of the project's data. ATLAScan creates a default image for each project.
• Custom preview image : Opens a browser window to let the user select a custom preview image file for
this project. The custom image file can be in the PNG, JPG, or TIFF formats.
• Last Modified: Date on which the project was last edited.
• Project Items: The number of particular items in the project, including point clouds, triangle meshes,
primitives and cameras.
• Location: Location of the project's files on the storage drive or network. Clicking Show in Explorer opens the
file location in a Windows Explorer window.
• Notes: A short user-defined description of the project, which is displayed in the Recent Projects window.
Clicking Edit Notes lets you edit the text field, clicking Cancel Edit removes your changes to the text field, and
clicking Save Notes saves the text in the text field.

Control
The Control environment lets you connect to and control a Polaris that is connected to the computer. The toolbar at the
top of the screen contains several tools for using the Polaris, while the status area at the bottom contains information
about the status of the Polaris. The region in between them contains different tools, depending on which function you
are using.
Toolbar
The toolbar contains the following tools for using the Polaris:
• Connect: ATLAScan opens the Connect window, which requests the IP address of the Polaris, then attempts
to connect to it.
• New Survey: Creates a new survey session.
• Manage Surveys: Opens the Manage Surveys window. This window enables you to download surveys from
the Polaris.
Status area
The status area at the bottom of the window contains the following information about the connected Polaris and its
recent activity:
• System Time: The time and date used on-board the Polaris.
• Icons: The status of the Polaris' subsystems, such as its battery condition.
• Open Log Window: Opens the Log Window. The text pane to the right shows the latest log message.

Control Window, Simple
The Simple tab of the Control environment lets you quickly specify the parameters for a new Polaris survey. For finer
control over the survey parameters, use the Advanced tab instead.
Contents
The Simple tab has the following parameters and tools:
• Presets: Selects a preset selection of parameters. Setting this to None enables you to enter manual
parameters below.
• Horizontal FOV: Sets the horizontal field of view within which the Polaris will survey.
• Vertical FOV: Sets the vertical field of view within which the Polaris will survey.
• Scan density: Sets the intended density of the point cloud. The Polaris adjusts the density by adjusting how
quickly the Polaris rotates and how quickly its scanner mirror oscillates. In Extra Coarse mode, they rotate and
oscillate quickly, which spaces the points far apart. In Extra Fine mode they rotate and oscillate slowly, which
packs the points closely together.
• Max Range: Set the maximum range that the Polaris will try to achieve. Note that this setting also affects the
pulse repetition frequency (PRF) of the Polaris (and therefore also its measurement speed), as shown below:
o If Max Range is 250 m, the PRF is 500 kHz
o If Max Range is 750 m, the PRF is 200 kHz
o If Max Range is 1600 m, the PRF is 50 kHz.
• Viewing pane: Displays a preview of the survey area taken during the preview scan. This is useful for setting a
custom FOV and performing target recognition.
• Survey Name: Sets the survey's name.
• Collect Images: Collects camera imagery. The camera images cover the same FOV as the lidar data. The
camera survey is performed after the lidar survey is complete.
• Start Scan: Begins the survey immediately with the current parameters.
• Polaris status bar: Displays system information from the Polaris.

Control Window, Advanced
The Advanced tab of the Control window lets you specify detailed parameters for a new Polaris survey.
Contents
The Advanced tab has similar tools as the Simple tab, but they allow finer control over the parameters.

Manage Surveys
The Manage Surveys window lets you view the surveys on the Polaris and download them to the computer running
ATLAScan.

File menu
This menu is organized into three main parts:
Project Management • New project: Opens the Create New Project Window, which lets you create a
new project and define its name, folder, and measurement units.
• Open project: Opens an existing project (an ASPRJ file)
• Save project: Saves the project under its existing name
• Save project as: Saves a copy of the project with a different name
• Open current project folder: Opens the current project's main folder in
Windows Explorer.
Import • Import: Imports data from third party formats.
• Import ATLAScan Project
• Load item
Export • Export model as…
• Export polyline as…
• Export cross section as…
• Export UCS to AutoCAD
• Export annotation to .CSV
• Export pose
• Export to Autodesk® ReCap 360...
Exit: Quits ATLAScan. If you have unsaved changes to your projects, ATLAScan asks you if you want to save the
changes or not before exiting.

Create New Project Window

The Create New Project window lets you specify basic information about the new project.
Fields
The window displays the following fields:
• Project name: The name of the project. This will be used for many of the project's files and folder
• Parent folder: The folder in which the project's folder will be created.
• Project file: The name and root of the ASPRJ file, based on the Parent folder and Project name fields.
• Change units: Opens the Manage units of measures window so that you can set the measurement units to be used
for the project.
Each project has its own folder within the parent folder. ATLAScan projects contain at least some of the following
subfolders:
• Imports
• Exports
• Unstructs
• Grids
• Meshes
• Polylines
• Trash
• Images
• Movies

Project Management
In this section you can learn how to efficiently manage your ATLAScan project by using the following commands.
New project
This function opens the Create New Project Window, which allows you to create a new ATLAScan project.
Open project
This function opens an existing ATLAScan project with the file extension ASPRJ. Before opening a new project, ATLAScan
asks whether you want to save the current one (if there are unsaved changes).
• When you open a project, ATLAScan loads all the project items in the project window, and sets the 3D scene's
current viewpoint to the position and orientation that it had when the project was saved.
• If you want to transfer or copy a project, you must copy the whole project folder with all its subfolders, not just the
ASPRJ project file.
Save project
This function saves the current project, safely storing all the project items, their loaded/unloaded state, their properties,
the current 3D camera pose, etc. It is advised to save the project frequently, especially before long processing
operations.
Save project as
This function saves only the ASPRJ project file with another name. This is useful for projects with many items, as it it
enables you to have one project with the point clouds, and another with the clouds and more accessory items like a
camera, planes or polylines.
You can also use this command to create a copy of the project: Save the project as a file in another directory, then copy
all the source project's subfolders into the destination project folder.

Import
To quickly import data, just drag and drop the files you want to import anywhere in the 3D scene. A wizard will appear to
ask you for import options, depending on the kind of data.
a. If you’re importing a point cloud the Process environment will be opened.

b. If you’re importing a different file format (as meshes, polylines, etc…) a wizard window will appear to guide you
through importing steps.
All imported models' coordinates will be interpreted in reference to the current UCS (User Coordinate System).
• Import Format
• Import Rec Project
• Load Item
See also File menu.

Import Formats
The Import menu item contains several types of data, such as grid point clouds, meshes, and more. Selecting a type of
data opens a window to let you browse for the desired file to import.
Grid Point Cloud

Comma Separated Values (CSV)
DEM ASCII header+matrix (ASC)
E57 structured point cloud (E57)
Generic binary raster (*.*)
Grid Point Cloud Text File (TXT)
Optech IXF scan (IXF)
Polaris point cloud (SCAN)
Polaris survey (SURVEY.XML)
Reconstructor RRF (RRF)
Triangle Mesh
3DS (3DS)
AutoCAD data exchange format (DXF)
Collada (DAE)
IFC, Industrial Foundation Classes (IFC)
PLY mesh (PLY)
Reconstructor VG3 mesh (VG3)
STL Mesh (STL)
Triangle Meshes Text file (TXT)
VRML (WRL, VRML)
Wavefront (OBJ)
Unstructured Point Cloud

E57 unstructured point cloud (E57)
Kubit PointCloud (PTC)
LAS format (LAS)
Optech CMS scan (TXT)
Optech XYZ format (XYZ)
PLY point cloud (PLY)
Unstructured Point Cloud Text File (TXT)
PTS (PTS)
Polyline/trajectory
Polyline
• AutoCAD data exchange format (DXF)
• List of points (TXT)
Trajectory
• Applanix SBET format (OUT)
• DXF trajectory (DXF)
• Optech LYNX cameras (TXT)
• Trajectory from point list (TXT)

CAD Models
AutoCAD data exchange format (DXF)
IFC (Industrial Foundation Classes) (IFC)
STL Mesh (STL)
VRML (WRL, VRML)

Import CAD models

This function allows you to import CAD software models into ATLAScan in order to do structural analysis or to compare
the models with the point clouds.
List of supported CAD formats

You can import CAD models in any of the following formats:
AutoCAD data exchange format (DXF) Imports meshes written in DXF format, ATLAScan will also read
their colors and textures.
IFC (Industrial Foundation Classes) (IFC) Imports models from software like Autodesk Revit via the IFC
format. ATLAScan preserves the model colors (except for the
transparency)
STL Mesh (STL) Imports STL files. ATLAScan asks for the measurement units in
which the STL vertices are written. Supported units of measure
are: meters, centimeters, millimeters, inches, feet, yards
If the STL file is binary, and if the string COLOR= is present in the
first 80 bytes, ATLAScan also tries to import the colors of the STL
solid. If it is not, no color is imported. The standard STL format
does not support colors or textures. This is only one of the non-
standard formats to write color into STL files
VRML (WRL, VRML) Imports VRML models as meshes. ATLAScan will ask you for a
scale factor, and will maintain the colors and textures of the
models.

IFC Import Dialog
Interoperability with Autodesk Revit

This import tool allows you to import your BIM project created with Autodesk Revit into ATLAScan. The BIM entities are
converted into meshes and can be compared with point clouds via the inspection functionality, to get accurate
deformation and structural analyses.
From Revit, just export your project in the IFC format, and then drag & drop your IFC file inside ATLAScan. The above
dialog opens up.
Exclude unneeded entities

BIM projects can be very large and detailed, including furniture elements, railings, windows, etc. For the sake of
inspection, you may want to exclude some project entities because, for example, they are simply not yet in place in
reality. To exclude some entity types, just mark them among the check buttons that represent the most commonly used
IFC entity types. In the screenshot above, for example, all furnishing elements, railings, ramp flights, sites and
BuildingElementProxys will be excluded from the import procedure. You can also specify which entity types to include ,
by selecting the Include only selected entities radio button. In this case, you have to check at least one entity type to
proceed importing.

Entities grouping
In the bottom box of the dialog, you have two grouping options: Group everything in a single mesh and Separate
entities by type into different meshes. The first option will merge all your IFC entities in a single mesh, resulting in a
faster inspection procedure. The second option will split your IFC entities among many meshes, one per entity type. If
you choose the second option, ATLAScan will create many meshes and put them under the same parent group in the
project. The meshes will be named using the following convention: <IFC file basename>_<IFC entity type>. For example:
“MyBuilding_IfcWall”, “MyBuilding_IfcRoof”, “MyBuilding_IfcWindow”, etc.

Import ATLAScan Project

This function imports items from another ATLAScan project. The items data can be copied or moved from the project to
be imported.

Load item
This function allows you to import an existing item saved in ATLAScan's internal format. The item is not duplicated,
copied or moved from its location, it is simply listed as belonging to the project.
With this function, you can load any of the following item types:
• Unstructured point clouds (RUP)
• Grid point clouds (RGP)
• Triangle meshes (RTM)
• Cross sections (RCS)
• Polylines (RPL)
• Trajectories (RTR)
• XML item file (XML)
These are all ATLAScan's internal formats for items.

Export
This menu contains a number of options to export ATLAScan's entities in various formats, to be imported in third-party
software or CAD tools for further elaboration.
• Export model as…

• Export polyline as…
• Export cross section as…
• Export UCS to AutoCAD
• Export annotation to .CSV
• Export pose
• Export to Autodesk ReCap 360
Note that when exporting models or other entities, the coordinates are exported as they appear in the current UCS.
See also File menu.

Export Model as...

This function allows to export a model (models are all point clouds and triangle meshes) into a third-party or
intermediate format for further processing in external software tools.
Export workflow
Use this workflow if you want to export a given list of models (clouds or meshes) to the same third-party format. After
selecting this function you are presented with a dialog where you can set format-specific options. Below is a table with
the supported options by format, and with the known limitations of each format. You can choose whether to specify
different export options for each model, or to adopt the same export options for all models. In the latter case, select Do
not ask again in the subsequent save filename dialog.
Format limitations when exporting global coordinates

Generally speaking, when exporting a model to an external format, ATLAScan tries to store as much information as the
format allows. Therefore, models are always exported in global coordinates, meaning the coordinates that the model
has in the current UCS. However, some format do not allow large coordinates for the models' points. Such formats are
STL meshes and PLY meshes. Because of intrinsic limitations of these formats, models exported in these formats are
saved in local coordinates, losing their global position in the ATLAScan project.
Export formats options and limitations
Grid Point Clouds
Export format Supported options Limitations

E57 grid point cloud (E57) Local or UCS coordinates
Exporting the reflectance in a way
E57 unstructured point cloud (E57) compatible with Autodesk ReCap.
Exporting a color layer
Generic binary raster (*.*) Exporting a scalar layer.
Grid Point Cloud (RGP)
Grid Point Cloud Text File (TXT) Object or UCS coordinates.
Subsampling factor for the exported
Unstructured Point Cloud Text File cloud.
(TXT) Number of decimals saved.
Skipping depth discontinuities
Skipping orientation discontinuities
PTC Kubit Point Cloud (PTC) Exporting a color layer.
[unstructured] Subsampling factor for the exported
cloud.
Skipping depth discontinuities.
Skipping orientation discontinuities.

LAS format point cloud (LAS) Local or UCS coordinates.

[unstructured] Exporting the reflectance in a way
compatible with Autodesk ReCap
Optech IXF scan (IXF)
[structured]
PLY point cloud (PLY) Local or UCS coordinates. If the model to be exported has
[unstructured] Exporting color layer extremely large UCS coordinates (more
Exporting point normals than 100 km), ATLAScan will not export
Storage mode (ASCII, binary little it in UCS coordinates due to PLY format
endian, binary big endian) limitations.
Coordinate precision (float or double)
PTX point cloud (PTX) Subsampling factor.
[structured] Exporting a color layer
Skipping depth discontinuities.
Exporting a PNG image of the selected
color.
Reconstructor RRF (RRF) Exporting a reflectance layer as
greyscale of a color.
Exporting the pose.
Unstructured Point Cloud (RUP)
PTS (PTS) Object or UCS coordinates.
Subsampling factor
Number of decimals saved
Unstructured Point Clouds

All the exported point clouds will be in an unstructured format.

E57 unstructured point cloud (E57) Local or global coordinates.
compatible with Autodesk ReCap.
PTC Kubit Point Cloud (PTC) Exporting a color layer
Subsampling factor
LAS format point cloud (LAS) Local or UCS coordinates.
compatible with Autodesk ReCap
Exporting a color

PLY point cloud (PLY) Local or UCS coordinates. If the model to be exported has
Exporting a color layer excessively large UCS coordinates
Exporting point normals (more than 100 km), ATLAScan will not
Storage mode (ASCII, binary little export it in UCS coordinates due to PLY
endian, or binary big endian) format limitations.
Coordinate precision (float or double)
Unstructured Point Cloud (RUP)
Unstructured Point Cloud Text File Object or UCS coordinates.
(TXT) Subsampling factor
PTS point cloud (PTS) Object or UCS coordinates
Subsampling factor
Meshes
3DS (3DS) Supports mesh textures It is not possible to save meshes with more
than 65536 vertices or more than 65536
triangles.
AutoCAD data exchange format (DXF) Does not support mesh color or mesh textures.
COLLADA (DAE)
PLY mesh (PLY) Exporting a color layer If the model to be exported has too big UCS
Local or UCS coordinates coordinates (more than 100 km), ATLAScan will
Exporting point normals not export it in UCS coordinates due to PLY
Storage mode (ASCII binary format limitations.
little endian, or binary big Does not support mesh color or mesh textures.
endian)
Coordinate precision (float or
double)
Reconstructor VG3 mesh (VG3)
STL mesh (STL) Does not support mesh color or mesh textures.
Only local coordinates are saved, the model
global positioning is lost.
Triangle Meshes (RTM)
Triangle Meshes Text File (TXT) Local or global coordinates
VRML (WRL, VRML) Supports mesh textures
Wavefront OBJ (OBK)

Export polyline
This option allows you to export a set of polylines to the DXF format to be used in other software (AutoCAD, for
example). When selecting this function, you are asked to select the output filename. After that, you are asked to provide
the decimal precision that the polyline's points should have.
You can specify a distinct output filename and decimal precision for each polyline. Otherwise, you can check Do not ask
again in the Select File dialog and you will only need to set the decimal precision once for all the polylines. The filenames
will be also assigned automatically.

Export cross section as

This option allows you to export a set of cross sections to the DXF format for use in other software (AutoCAD, for
example). When selecting this function, you are asked to select the output filename. After that, you are asked to provide
the decimal precision that the cross sections' points should have.
You can specify a distinct output filename and decimal precision for each cross section. Otherwise, you can check Do not
ask anymore in the select file dialog and you will set the decimal precision only once for all the cross sections. The
filenames will be also assigned automatically.

Export UCS to AutoCAD

This function exports an User Coordinate System to AutoCAD. The UCS is saved as an SCR file, a script that can be loaded
in AutoCAD.
To import an UCS in AutoCAD, type _script, press Enter, and select the script file to open it. The imported UCS is set
as current and the top view is applied.

Export annotations to CSV

This function exports a set of point annotations obtained with the annotation tool into the CSV (comma-separated
values) format, which can be read by spreadsheet software like Microsoft Excel.
Select a list of annotations from the project window, select this function, and enter a file name to export them. In the
CSV file, your annotation data will be organized in six columns, and as many rows as there are annotations:
• The first column contains the annotations' unique IDs.

• The second to fourth columns contain the annotations' Cartesian coordinates.
• The fifth column contains the comments.
• The sixth column contains the annotations' hyperlinks, if any.

Export pose
This function stores the positions and orientations of your project items in a separate folder so that you can recover
them if needed. This is a very useful function for registration workflows.
When pre-registering, registering, geo-referencing or moving around your models, you may want to backup the model's
positions and orientations (poses) before changing them. If something doesn't work in the registration algorithm or in
changing the models' poses, you can safely restore the poses as they were before the action.
When you activate this function, the poses of the selected items are saved in the Exports project folder, in a sub-folder
with the name pattern PoseFiles_yyyy.mm.dd_hh.mm.ss_xxx, which indicates the precise date and time when the
poses were created.
To restore any item's pose, use the pose transform dialog.

Export to Autodesk® ReCap 360 Pro

This function directly exports data in RCS and RCP formats for Autodesk® ReCap 360 Pro, enabling a more complete and
faster integration with that software. ATLAScan point clouds (RGP and RUP) can also be directly imported in ReCap 360
Pro.

Navigation menu
ATLAScan provides many interaction styles to navigate in the 3D scene. The navigation modes are selectable through the
Navigation toolbar (placed on the right) or by Navigation command in the menu bar.
Predefined views
Navigation Modes
Screen settings
Creation of User Coordinate System
Other commands are available In the context menus of certain items:
Go to
Centers the 3D window to the bounding box of the model. The view point is computed so that to contain the
whole bounding box in the viewport. This tool is very useful when you have loaded many objects and want to
go quickly and easily to a specific one. Warning: This option may place the model far away, with the effect that
it seems not visible. In this case, increase the max depth in the options.
Center to local origin
Aligns the view point to local coordinate frame of the model, with X rightwards, Y upwards and Z going out of
the screen.
Align to bounding box
The view point is aligned to the desired face of the bounding box of the model.
In View Parameters dialog you can customize advanced parameters of the current view by controlling how the “virtual
eye” sees the 3D scene with all your models.

Predefined views
Default view
This navigation mode allows the user change the current view so that all the selected items are completely visible in the
3D scene.
Bubble View
This navigation mode allows the user to place the current view to the center of an item. Only camera rotations are
allowed in this mode. See also Bubble mode.
2D View
This navigation mode opens the Edit 2D Grid Window. Note that this view is supported for grid point clouds only.
Map View
This navigation mode places the view so that all the selected items can be seen from a top view.
UCS Views
This navigation mode places the view so that all the selected items can be seen from an according to UCS view:
See also here.
Recent camera positions

Cycles through recent camera positions.
Other predefined views are available in the Process environment:

• Slice View
• Horizontal Section View
These modes can be used to check the correct alignment along a vertical/horizontal section of the displayed items.

Slice View (Process Environment)

This navigation mode automatically switches the view to orthographic mode and aligns it to the selected
items bounding box vertical plane.
This mode can be used to check the correct alignment along a vertical section of the displayed items. The depth of the
slice can be changed from the options dialog.
Keyboard/mouse movement Effect

Rotate the around the selected
LMB + move mouse
items vertical (z) axis
CMB + move mouse Translate sideways
Translation towards point indicated

Mouse wheel rotation
by the mouse

Horizontal Section View (Process Environment)

This navigation mode automatically switches the view to orthographic mode and aligns it to the selected
items bounding box horizontal plane.
This mode can be used to check the correct alignment along an horizontal section of the displayed items. The depth of
the horizontal slice can be changed from the options dialog.

Rotation the around the selected
LMB + move mouse
items vertical (z) axis
CMB + move mouse Translation sideways

by the mouse

Navigation modes
Pivot mode
Default navigation mode based on translating and rotating around a chosen point of the 3D scene
Walk mode
Navigation mode that allows you to translate and rotate around yourself, independently of the
models
Bubble mode
Navigation mode in which only rotations around yourself are allowed
Zoom
Narrows down the virtual camera's field of view to concentrate on a particular viewport. It doesn't
move the virtual camera
Move to viewport
Moves the virtual camera so that a particular viewport is in sight, without changing the field of view

Recovers up to 1000 steps you went through during your navigation

Pivot Mode
In this mode a point can be selected as a rotation pivot. The pivot is set by a left button click on a valid point.

LMB + move mouse Camera rotation (pan and tilt)
Camera sideways translation (tracking
CMB + move mouse
and booming)
Camera dollying forward and
RMB + move mouse back/forth
backward along the current focal axis
by the mouse
Shift + LMB + move mouse left/right Camera rolling
LMB click on a model's point Selection of rotation pivot
Zoom in and out (change of field of
Shift + mouse wheel rotation
view)
Selection of points for point list
Alt + LMB double click
window
Shift + CMB + mouse move Draw rectangle to move into
Copy readout window's contents to
C button on keyboard
clipboard

Walk mode
This mode is designed to simulate a human being walking around the 3D scene. The user can make translation
movements with the keyboard and mouse movements with the mouse.

LMB + move mouse Camera rotation
w, x, a, d keys Move forward, backward, left, right
q, z or +, - keys Move up, down

Bubble mode
LMB + move mouse Camera rotation (pan and tilt)
Mouse wheel rotation Zoom in and out (change of field of
view)
Shift + LMB + move mouse left/right Camera rolling
Alt + LMB double click Selection of points for point list
window
C button on keyboard Copy readout window's contents to
clipboard

Zoom Tool
This tool allows the user to zoom on a specific part of the scene.
The tool can be used by following these steps:

• Press the Zoom command in the Navigation menu.
• Drag the mouse while pressing the left mouse button to draw a rectangle in the 3D view.
• Release the mouse: The camera zooms into the new defined viewport. The system then changes back to the last
navigation mode used.

Move to viewport
This tool moves the viewport to a specific part of the scene.
The tool can be used by following these steps:

• Click Move to viewport command in the Navigation menu.
• Drag the mouse while pressing the left mouse button to draw a rectangle in the 3D view.
• Release the mouse: the camera translates the new defined viewport. The system then changes back to the last
navigation mode used.
Difference between Zoom tool and the Move to viewport tool

• With the Zoom tool, the camera is zoomed-in on a detail of the scene. Only the field of view (in perspective mode),
or the projection size (in orthographic mode) is changed.
• With the Move to viewport tool, the camera is translated closer to the detail indicated by the rectangle, leaving the
field of view unchanged.

This dialog enables you to search through your most recent 1000 camera positions. If you make a mistake, lose your
models, or lose the position you reached when navigating your models in the 3D view, you can use this function to
retrace your steps.

Screen settings
Show axis
Shows/hidesthe coordinate system frame of the current camera global reference (CCGR) in the bottom left corner of the
3D scene.
Show bounding box

Shows/hides the bounding boxes of all items in the 3D scene.
Color by ID
Renders all the project entities with their identifying color ID. The color ID of a point cloud can be changed from the
property window. This option is useful for distinguishing the models from each other and to know which model is which
in the 3D scene.
Options
Modifies the current navigation and rendering parameters.

Display and navigation options
This dialog is accessible by selecting Options from the Navigation menu, or by selecting Options in the Screen
Settings menu in the toolbar menu.The options present in this dialog are described below.
Enable human movements while rotating

When this options is checked, a small human avatar appears in the bottom-left corner of the 3D rendering window.
While you navigate in the 3D scene, you will experience that after doing many rotations, your viewing direction will still
be oriented according to the vertical axis and to the horizontal plane defined by the current UCS. This is like a human
being that, after rotating his head around his neck in many directions, when relaxing the head will go back into the
natural position, which is in line with the spinal chord.
When this option is unchecked, while navigating the 3D scene you will experience that after some rotations the
reference to the UCS's horizontal plane is lost, and that you are floating in the space like an astronaut floats among
meteors, without gravity force.
Use scalable quick rendering

This option turns on scalable quick rendering. When hundreds of millions of points must be rendered, this rendering
system allows you to still navigate very fast across your models. This is possible by rendering only 500,000 representative
points of all the loaded point clouds. One after second you stop navigating, ATLAScan starts rendering the complete set
of points.
If this option is unchecked, ATLAScan will render the complete 3D scene for each refresh of the 3D window, up to the
defined maximum number of rendered points. In selection mode, the scalable quick rendering is automatically disabled.
Set max. rendered points

This option is used to set a maximum number of point to be rendered in the 3D scene. The limit is cumulative for all the
loaded point clouds.

Set max. rendered points in Virtual Scan

This option is used to set a maximum number of point to be rendered in the Virtual Scan tool. The limit is cumulative for
all the loaded point clouds.
Dynamic point size

If this option is enabled, the size of the rendered 3D points decreases linearly with the increasing distance of the points
to the “virtual eye”. If this option is disabled, the size of the rendered 3D points is constant regardless of the distance
from the viewing point.
Point smoothing
Enable this option to render 3D points as small circles, disable it to render 3D points as small squares.
Projectors cast shadows

If the graphics card supports shadow mapping, this option allows to render projectors that cast shadows through the
scene geometry. For best quality, render only triangulated models. Although this can be done also with point clouds by
increasing the point size, color projected by the projector could “bleed through” the occluders.
Use shaders for projectors

If the graphics card supports CG shaders, this option can be enabled, which allows to render projectors with the highest
quality. Nvidia graphics cards always support CG shaders, while ATI graphics cards don't.
Blend ColorID with point color intensity

If the point clouds are set to be colored by ID, each point is shaded according to its intensity value. High-intensity points
are lighter shades, while low-intensity points are darker shades. If this option is not selected, all the points have the
same shade.
Draw annotations to 3D scene sides

Moves the text boxes for all annotations to the right edge of the 3D view. If this is not selected, the text boxes are kept
near to their associated object.
Movement speed
This option sets the translation speed when panning in pivot mode.
Slice views thickness

This option sets depth of a single slice when using the slice view or the horizontal section view.
Set gradient background

Change the uppermost and lowermost color of the background of the 3D rendering window.
Object reference frame size

All models and cameras can draw their local coordinate frame by enabling the property. Here the frame size (in meters)
is specified to improve its visibility accordingly to the extent of the scene.

Pivot mode controls

Here you can select two options: Reconstructor and Move with 3-button mouse. These options are two sets of controls
for the Pivot Mode. The second one is the default one and is described here.
If you select Reconstructor, then in pivot mode the following controls apply:
• Rotate: To rotate the model around the pre-defined centre of rotation (pivot), move the mouse within the 3D
window while keeping the left button pressed.
• Pan: To pan the model in X and Y, move within the 3D window while keeping the left button and the Shift key
pressed.
• Zoom: To translate the model along the Z-axis, move the mouse within the 3D window while keeping the left
button and the Ctrl key pressed or use Ctrl + mouse wheel to zoom in and out along the Z-axis. In the latter
case, if the mouse pointer hovers a valid 3D point, the zoom converges to that point.
• Change pivot: To change the pivot, double-click the left mouse button plus Ctrl+Shift on a valid 3D point.
Bounding box colored with colorID

When this option is checked, the bounding boxes of the project items are colored based on the item's colorID. If this
option is unchecked, the bounding boxes are colored with the default (yellow) color.
Bounding box edge smoothing

Apply an anti-aliasing algorithm to bounding box edges. It is suggested to leave this option checked unless the graphic
card does not support lines smoothing.

View Parameters Dialog

This window appears when you click the View parameters button in the top toolbar. Here you can customize advanced
parameters of the current view by controlling how the “virtual eye” sees the 3D scene with all your models. There are
two viewing mode: Perspective mode (left) and orthographic mode (right).
Perspective projection
In perspective mode, the virtual eye uses a perspective projection to see the scene. You can set the amplitude of the
projection by specifying Vertical FOV (field of view) in degrees. Decreasing or increasing the field of view is equivalent to
zooming in or out (respectively) with the objective lens of a photocamera.
Orthographic projection
In orthographic mode, you see the 3D scene through an orthographic projection. You can specify the Vertical size of your
projection in meters.
Cut planes
In both perspective and orthographic mode, you can adjust the cut planes:
• Near plane: Distance of the near clipping plane in meters. Specifies the minimum rendered depth. In perspective
mode, only values greater than zero are allowed.
• Far plane: Distance of the far clipping plane in meters. Specifies the maximum rendered depth. Must be greater
than Near plane.
To optimize the depth accuracy, try to keep these values as close as possible to the desired scene depth range. For
example, if your model's bounding box is 100×100×100 m, it is not useful to adopt a far plane of hundreds of kilometers.

Create UCS from point lists

This command creates a new UCS from two sets of points. Switching to the resulting UCS will show the points
of the first set with the coordinates of the second set.
This function is accessible from the Navigation menu, or from the toolbar of the point list window.
Given two lists of points, this function creates a new UCS such that the points having coordinates in list 1 will assume the
coordinates in list 2 after switching to the resulting UCS, except for a controllable mean-squared error.
Click Load moving points to load a text file with the first point list, then click Load reference points to load a text file
with the “reference” coordinates of the points (the coordinates that are going to be shown in the new UCS). To parse the
text files with the point list, an appropriate tool is used. You can use the points listed in the point list window as moving
points by clicking Copy moving points from point list window.
After loading the first and the second list of points, click Compute. A dialog appears, asking whether you want to register
the points by coupling them according to their labels, or by trying out all the possible pair combinations to find the best.
The first option is much faster but it assumes that you are sure about how to match your points. After you have selected
either Match names or Best fit, you can refine and make use of the results of the registration, in the dialog shown
below.

In the dialog above, it is possible to set the error threshold to exclude outliers in order to improve the mean registration
error, as described for the ICP Registration Dialog. Once you have refined the mean-squared error (shown in the top
left), click Create UCS from transform to create the desired UCS. The resulting UCS will be inserted in the project
window.
This procedure uses the same calculations as georeferencing from the point list.

Outputs menu
This menu contains many functions to extract results from the project's data and models. These functions are organized
in the following sub-menus:
• Measures & notes

• Elevations & plan
• Cross sections
• Areas & volumes
• Inspection
• Geometric shapes tools
• Tunnel survey
• Video record
• Drawing tools
• Save snapshot
• Manage units of measure

Measures & Notes

ATLAScan features the following basic measurement features:
Readout The readout window allows you to see the 3D coordinates of the points you are hovering
window the mouse on in the 3D rendering window.
Distance This button allows you to measure a distance between any two points in the 3D scene.
After activating this button, left-click on the first point, drag the mouse to the second
point, then release the mouse button. A result dialog will pop up displaying the distance
between the two points. You will be offered the option to save the distance as an
annotation in your project.
Angle This button enables you to measure and angle between three points in your 3D scene.
After activating this button, click the mouse's left button (LMB) for each of the 3 points
(which must be valid 3D points). The angle of the 2nd point from the 1st and 3rd is
computed, and a result dialog opens up to display the result in degrees. You are also
presented with the option to save the angle measure as an annotation in your project.
Annotation This function allows you to create annotation of 3D points of models in your project. After
activating this button, double-click with the left mouse button on a valid 3D point to place
an annotation. The annotation gets displayed in 3D and appears also in the Annotations
folder of your project. You can edit the annotation's comment and set an associated
hyperlink in the Property Editor.
Exporting meaningful points to Excel or other software

Using this function, you can annotate the meaningful points of your project (e.g.
topographic targets) and save these annotation as project items. Successively, you can
export these 3D annotations in CSV format and import them in Excel or other external
tools for further processing. See also the Basic measures menu.
Open This button enables you to measure the angle between different objects in the 3D view.
Angle Tool After clicking the button, a pane opens showing the options to measure the angle between
any of the following: three points, two segments, a segment and a plane, and two planes.
After selecting the type of object to measure, click on the specific objects in the 3D view.
See also Areas & Volumes.

Elevations and plans

This sub-menu contains all the commands and features to create precise, high-resolution elevations and plans of your
models:
• Create orthophoto
• Create camera
• Place here survey point
• Virtual scan

Create Orthophoto
With this toolbox you can create an orthocamera from samples, points, axes, UCS directions, etc., edit it and extract an
orthophoto through the Virtual Scan tool.
The recipe window is split in two panels:

• Position
• Dimensions and corrections
Position
The panel of the dialog allows you to create an orthocamera from selected samples using the Quick selection tool (as in
Selection tools) or from a point list. Clicking Fit Orthocamera creates an orthocamera. It is then possible to edit the
camera, align it to UCS views, and make it horizontal or vertical.
At the end of this process click Done, set camera dimension to switch to the Dimension and Corrections panel.

Dimensions and corrections

Here you can set the properties of the camera: dimensions, resolution, near and far plane distance. By using the shift
command you can translate and rotate the camera in the space, as you see it in the 3D window. The Orthophoto
Preview button helps you to find the right position of the camera by showing you a preview of the final result.
At the end of this process press Done, snap Orthophoto button to open the Virtual Scan window.
See also Virtual scan and Exporting orthophotos to AutoCAD.

Create Camera
This submenu creates a camera with its origin in the current point view. It can create several types of camera:
Perspective
Orthographic
Spherical
Cylindrical
Note: Perspective Camera

A perspective camera can be used to model a common photographic camera (e.g. a reflex camera), which follows the
rules of perspective projection. This type of camera can be used to acquire a virtual scan or to define a point of view in
the scene, which can be fixed by right-clicking on the camera item inside the project tree and selecting Apply projection.
This camera can be converted into a perspective projector by setting an image using the Set projector image command.
Note that the image to be applied needs to be undistorted.
See Cameras for more informations regarding perspective cameras properties and associated functions.

Place here survey point

This tool creates a perspective camera with its origin at the current point view. Its properties are the same as the
current settings in the View Parameters window.

Virtual Scan
This function is named "Virtual Scan" because you can imagine it like a scanning of the loaded models in the 3D window
with a virtual laser scanner. It's useful for resampling the 3D dataset on a 2D grid from a defined position in 3D space and
with a defined projection (orthogonal, perspective, cylindrical, spherical). Check the models that must be scanned in
the Project Items Window, and the depth range of the scan (near and far clip planes) of the current camera. The virtual
scanner can create snapshots, cube maps or grid point clouds, and also does texture blending and mapping for
triangulated models.
The following options are displayed:
• Do virtual scan: Creates a preview of the scan from the selected view point and projection. Select the resolution and
background color first.
• Pick mode:
o Distance: Left-click on the first point, drag the mouse to the second point, and release the mouse button.
Both endpoints must be valid 3D points.
o Point: Double-click to pick a list of 3D points from the virtual scan. If the AutoCAD link is enabled, the points
are sent to AutoCAD. Check the Global Coordinates for reading global/local coordinates.
• Save grid point cloud to project: Creates a grid point cloud from the current virtual scanned color and depth. The
model is automatically added to the project.
• Save image: Saves the current virtual scanned image to file.
• Save cube map: Use this view point to construct a cube map of the scene, i.e. 6 square images (the faces of the
cube) of size Width with perspective projection (90° vertical field of view) are generated around the centre of
projection and saved to file
The bottom row of the dialog shows the unit scale (depends on the type of projection) of the image and the computed
volume between the plane of projection and the visible 3D data. For best results use triangle meshes, so no holes are
found on the surfaces. Otherwise, try to increase the point size of the cloud of points in the Property Editor.

Orthophotos
There are several extra options if the camera is a plane/ortho camera. A virtual scan of a plane/ortho camera or
projector will be an orthogonal projection and thus an orthophoto. If the image is saved with Save image, a text file is
created along with the image file which exports registration information of the ortho image in the scene.
When ATLAScan saves the image, it also generates a script for AutoCAD®. This allows you to load the orthophotos in the
correct position and scale (in a blue layer with the same name dedicated to image) by simply dragging and dropping
them in the AutoCAD® window. It also creates a UCS having XY drawing plane coincident with the image plane and
positioned in the lower-left side of the orthophoto, in order to facilitate redrawing the vectorial draw.
Press the Export to AutoCAD settings button to choose the options. Remember that the DWG file must be saved in the
same folder as the script files (SCR).
This text file can also be imported in AutoCAD® with the Kubit® plugin.
Gridding: This process computes the optimal estimation for missing points in a grid of size MxN, given K valid initial
points. The algorithms available are: simple kriging (user has to specify the global mean), ordinary kriging (the mean is
computed automatically), kriging with trend (the mean varies smoothly). Gridding is a demanding process, with the time
complexity being O(K²MN), so use as few as possible initial points.
See also Create Orthophoto.

Exporting orthophotos to AutoCAD
Several options are available to export your orthophoto to AutoCAD.
• Export 3D position and orientation of orthophoto to AutoCAD: Exports the photo's current 3D position and
orientation in ATLAScan's UCS to AutoCAD. If this is not unchecked, only the plain 2D image is exported. Exporting
the image's 3D pose to AutoCAD is useful if other items (e.g. polylines) related to the same model are exported to
AutoCAD.
• Create new UCS in AutoCAD with same position and orientation of orthophoto: Createsa new UCS with the same
position and orientation as the image in AutoCAD. This checkbox is only enabled if the previous box is selected.
• Create a new layer in AutoCAD containing the orthophoto: Exports the orthophoto in a new AutoCAD layer. If this
is not selected, the orthophoto is exported in the current layer. Depending on the number of layers in the AutoCAD
project, you may want to add a new layer or not.
• Export dimension style: Exports all parameters of a dimension style, such as font size, color, arrow size, etc.

Cross Sections
This submenu contains commands and features to create cross sections and isolines of point clouds and meshes:
• Create/edit plane
• Cross sections
• Cross sections from planes collection

Cross sections
To create cross sections of your models, select Cross sections > Cross sections from the Outputs menu, or
right-click on the plane and select Cross sections > Cross sections from the context menu.
You can make cross sections of grid point clouds and meshes. However, it is not possible to make cross sections of
unstructured point clouds.
Cross sections are defined with respect to a plane.
With the default settings, the section is defined as a polyline created by cutting all loaded models with the plane.
However, it is also possible to have multiple cross sections by defining more slices or additional cutting planes, all parallel
to the original plane and lying at a certain distance from it.
Number of slices specifies how many slices are to be produced (the default is one). Slice spacing specifies the default
inter-slice distance. When you edit these two parameters, the values in the panel on the right in the dialog change
automatically. These are the list of the slices' distances from the plane. By double-clicking on a single distance, it is
possible to edit individually each distance. Furthermore, it is possible to add a slice by clicking , or remove a slice by
selecting it and clicking . There are also two buttons to Save the distances and Load the distances: these are useful
when you have to use the same distances set for many cross sections.
Distances are defined on the plane's Z axis, therefore a distance of 0 means that the the slice is lying on the plane, and

the positive direction is given by the direction of the plane's Z axis. You can invert the plane's direction with the
appropriate command in the plane's context menu. It is also possible to create negative distances by double-clicking on
them and writing a negative number.
Towards the bottom of the dialog there is the Group polylines by panel, which allows you to group the output cross
sections in different ways. Selecting Slice creates one cross section for each slice, regardless of how many models the
slice crossed. Selecting Model creates one cross section for each model crossed, regardless of how many slices crossed
that model and where.
When you click Ok, the specified cross sections are computed and added to the project under the Cross sections items
group. Cross sections can be exported in the DXF format for AutoCAD.

Cross sections from plane collection

This command creates a list of cross sections starting from a sequence of planes.

Area & Volume

These tools compute area and volume of point clouds and meshes portions.
Area calculation
Volume calculation
Compute cut & fill volumes

Area
ATLAScan provides different ways of computing areas.
When you click the Area button you can choose if you want to compute an area defined by a mesh or by a point cloud.
1. Area of a mesh portion

By using the Sampling tool you can perform a selection of a mesh's portion. A new mesh is created using only the
geometry visible from the current view point.
2. Area of a point cloud portion
• Sampling tools: This selects the point cloud portion using the sampling tools on the 3D window. This selection can
span across multiple point clouds.
• Polyline: This selects the point cloud portion using the Point selection with polyline command.

Volume
This process enables you to calculate volumes by integrating meshes, or portions of meshes on a custom plane. You can
activate this tool via Outputs > Area & Volume > Volume in the main menu, or via the same button in the top toolbar.
On activation, the window below appears in the Recipe window.
You can define the mesh, the plane and optionally the polyline to use with the usual drag-and-drop mechanism.
ATLAScan calculates the volume of the mesh integrated onto the plane, regardless of the triangles' normal directions. It
should be noted that only the portion of mesh lying in the positive semispace of the plane will contribute to the volume.
Parts of mesh lying below the plane will result in zero volume.
Optionally, you can also provide a polyline. The polyline, projected on the plane, defines a closed polygon on it. If you
provide a polyline, only the mesh portion whose projection falls inside that polygon will count for the volume, as if you
had cut the mesh with the polyline before computing its volume.
There are three ways to define the reference plane: using the current view, dragging a custom plane, or using the

horizontal plane with the given altitude on the current UCS.
You can compute the volume above the plane (and under the mesh) or compute volume below the plane (and above the
mesh).
If you check Preview volume region, ATLAScan will show a preview of the 3D region from which the volume will be
computed. This preview refreshes if you change any of the parameters from inside the recipe window. It may not refresh
if you change something from outside the recipe window, such as if you move the plane or the mesh with the Adjust
pose tool.
When you press Ok, ATLAScan displays a progress bar showing the calculation's progress. When it finishes, it displays a
short report on the volume measure. You have the options to copy this short report to the clipboard, or to
access another window that enables you to produce a detailed PDF report, via the Get PDF report option.

Volume PDF Report

This dialog enables you to create a detailed PDF report about your volume calculation.
The dialog allows you to fill in a series of fields with the information you want to appear in the report. The fields you
uncheck or you left blank will not be included in the report.
On top left of the dialog you can fill the fields Site name and Company name. If filled, they will appear as title/subtitle of
your report. Below, check Survey information if you want to include in the report information about the survey operator
and survey date. The checkable panel Processing information below works in the same way.
On the top right of the dialog, you can also select whether you want the current 3D view to appear as a screenshot in
your report. Finally, on the bottom right you can type additional notes to be included in the report.
The data you insert are saved also in Windows' registry, therefore the next time you use this dialog you will be able to
reuse the data you filled in before.
When you have inserted the data of your report, click Save PDF report and select a filename. The PDF file you specified
will be generated.

Compute cut & fill volume
This procedure implements calculation of cut and fill volumes. Given two meshes representing the same object (e.g. the
same terrain) at different instants in time, the cut volume is the volume that the object lost between instant 1 and 2,
while the fill volume is the volume that the object gained between the two instants.
The simplest way to use this procedure is to drag in the recipe window one mesh representing the surface at time 1,
another mesh representing the surface at time 2, and click OK. ATLAScan will return the volume of surface 1 integrated
on the Z=0 plane of the current UCS, the volume of surface 2 integrated in the same way, the cut volume (by how much
the surface was excavated between time 1 and time 2), and the fill volume (by how much the surface was filled (has
gained volume) between time 1 and time 2).
Given two surfaces S1 and S2, and a plane P, and assuming that the projections of S1 and S2 on P share a support C in

common, a third surface Smin can be defined. Since each point in C, is the projection on P of (at least) one point
of S1 and one of S2, the one of these points that is closest to P is defined as belonging to Smin. Let V1 be the integral
of S1 on C, V2 the integral of S2 on C, andVmin the integral of Smin on C. Then the cut volume Vcut is given by V1 - Vmin,
and the fill volume Vfill is V2 – Vmin.
The user can specify several optional parameters, to satisfy special requirements. A set of altitudes can be optionally
defined, ATLAScan will calculate the contributions to cut and fill volumes for each bench among two neighboring
altitudes. The altitudes can be added and edited manually. Clicking adds a new altitude, while removes the selected
altitude. Double-click on an altitude to change its value. Otherwise, the altitudes can be generated automatically by
checking the Automatic creation of regular benches and specifying a starting altitude and a fixed bench height.
These altitudes are by default referred to the Z axis of the current UCS. However, the user can change this by dragging
and dropping a plane into Reference plane (optional), which becomes the reference of all cut and fill calculation and of
all altitudes.
The last optional parameter is a polyline that the user can specify to delimit the support C on which the surfaces are
integrated. The closed polyline is projected on the horizontal plane Z=0 of the reference system and the support C in
common between S1 and S2 is intersected with the area delimited by the projected polyline. Therefore, the volumes are
calculated only inside the frustum defined by the polyline and the horizontal plane.
Once you have set all the parameters, click Ok. The computation starts, and the status is shown by a progress bar and by
messages in the log window.
See also the Cut and fill by altitude report.

Cut and fill by altitude report
This dialog provides you a report of the results of a cut and fill calculation, along with options for managing and
exporting the results.
From top to bottom, the names of the “before” surface and the “after” surface are reported, then the volume of the
surfaces, then the cut and fill volumes, then the number of benches considered for the calculation. At the bottom is a
table with the results displayed for each bench. You can define benches (intervals among altitudes) and see how the cut
and fill volumes behave for each bench.
On bottom right, four buttons offer you options for exporting the results. You can copy the results in text form to the
system clipboard, save the result to an XML file, or create a PDF report with the result. This report can be integrated with
the company name, site name, survey information, and processing information. Generate PDF report opens the Volume
PDF report dialog.
You can also Change units of measure displayed in the report.

Inspection
The Inspection procedure compares two different shapes and measures their differences. It is normally used to compare
two models of the same objects surveyed at different moments in time. For example, it could be used for monitoring
barrels of nuclear waste to immediately detect possible deformations.
The Inspection procedure is accessible by selecting Outputs > Inspection in the top menu bar, from the corresponding
button in the top toolbar, or from the context menu of any point cloud (structured or not) and any mesh.
The result of the inspection is a new color layer that will be assigned to the point cloud. For each point, this color layer
carries its measured distance from the reference model. After performing Inspection It is possible to use the Colors
mapping dialog function to map these measures with colors chosen by user and to see the corresponding color
scale and Histogram.

Input models
As input, user must provide a model to inspect and a model to use as a reference. The model to inspect must be a point
cloud. The reference model must be a triangle mesh.
User can input the desired models by simply dragging and dropping them, or by clicking Load model to inspect and Load
reference model. Click Remove to deselect undesired models from your import.
Parameters
At the center of the dialog there is the Parameters box that takes:
• Max search distance: every point of the model to inspect will be compared with the reference model only inside a
sphere with a radius defined by this parameter.
• Subsampling (factor s): Adjusts the speed of the calculations versus their precision. The exact effect of this factor
depends on the type of point cloud being inspected:
o Inspecting an unstructured point cloud: The cloud will be divided in s sets of points and only one point in
each set will be inspected. The other s – 1 points will be assigned the same distance value measured for
that point.
o Inspecting a grid point cloud: The procedure will scan the points according to their structure. The grid's
structure will be divided in cells of dimension s by s. Only one point per cell will be inspected and the other
points will get the same inspection value measured for the first.
• Inspection name: Defines a name for the layer that will be that will be generated
Inspection types
User can select the Inspection type to be performed by clicking Change Inspection type button.

There two main inspection types, each with its down options:
• Complete Inspection: With this method, you can select Signed values or Absolute values.
o Signed values: Distances will be computed as 3D vectors.
o Absolute values: Distances will be computed as absolute values.
• Inspection along one direction: With this method, you can select Cloud Origin->Point or one of the UCS axes from
the dropdown box.
o Cloud Origin->Point: This is much faster than a complete inspection, but also less accurate. For each point
in the input cloud, ATLAScan casts a ray from the cloud's origin and computes the location where it
intersects the reference surface. Finally, ATLAScan calculates the distance from the point along the ray.
Since this procedure only looks in one direction, it can measure an infinite distance for a given point, even if
the reference surface is very close to the point along another direction.
o UCS axes: This procedure takes each point of the cloud and computes its distance from the reference
model along a specific axis. User can choose one of the three axes of the current UCS. This inspection type
is faster than a complete inspection, is usable with a large value for Max search distances (e.g. 50 m) and
provides signed distance values (signs are calculated using the normals of the reference mesh's triangles).
Clicking Process starts the inspection. The resulting layer is appended to the color type list of the inspected point cloud,
using the specified Inspection name. Use Colors mapping to optimize the display with a legend.

Geometric Shapes Tools

This menu item contains the following tools:
• Create/edit plane
• Create/edit segment
• Create/edit circle
• Create/edit polyline
• Create/edit sphere
• Create/edit cylinder
• Create/edit cone
• Distance between shapes
• Intersect geometric shapes
• Edit points list
• Polyline from point list
• Flythrough from point list

Create/edit Plane
This window is a toolbox for creating/editing your plane from points, axes, objects, UCS directions, etc. To create a new
plane, click the Create New button ( ) To edit an existing plane, drag the plane into the Drop a plane here box, or click
the Browse button ( ) and select it from the list. The perimeter and area of the plane are shown below. The following
settings are available for creating or editing the plane.
Parameters
This setting defines the plane's dimensions along its X and Y axes.

Editing method
There are several ways to define the plane, such as Fit plane to samples and Fit plane to point list. Selecting one of the
options causes the window to display tools for creating the plane using that method.
Fit plane to samples
In this method, ATLAScan creates a plane based on a large number of points that you select in the 3D view.
Fit plane to point list
In this method, you can create a plane from, or to make your plane pass by, one to three points. These points are the
center of the plane, the X axis endpoint, and the Y axis endpoint. To define one of these points, just click on the
corresponding select button and Alt + double click on any point in the 3D scene. Otherwise, select any point from
the point list window. When you have defined the point(s), click on create/edit from specified points. The plane will be
adjusted to pass by the specified points, leaving as much as possible unchanged the other plane's properties.
From segment and point
In this method, you create a plane based on a single segment and a point.

From origin/axes
In this method, you create a plane based on the UCS origin, the X axis, or the XY plane.
From UCS axes layout
The buttons in this panel adjust the plane's spatial orientation. The first button makes the plane vertical in the
current UCS, without moving the central point of the plane. The second button makes the plane horizontal. The third
button rotates the plane around its normal vector (passing by its center) until the plane's X axis is parallel to the XY plane
of the current UCS.
From bounding box of selected items
The lowest panel of the dialog allows you to create a plane from the faces of the cumulative bounding box of the
currently selected project items. This is useful for example to immediately create the base plane of a certain church or
cave, to make plants or sections later.
When you create/edit a plane from the bounding box of project item(s) (bottom group box), the plane's direction is
determined by the current UCS: frontal, lateral and base plane mean that the plane's normal is parallel to the UCS' X, Y
and Z axis respectively.

Duplicate plane as
When you have created or edited the plane, you can save it as a new plane or as a new orthocamera. If you are simply
editing an existing plane, you can just close this window and it will be saved automatically.
Tips and tricks

The functions explained before are more powerful if used in combination. For example, first you create a plane from
three points of a façade, then you make the plane vertical, or its X axis horizontal, to create later an orthocamera from it.
Note also that, while you edit your plane, you have the undo/redo buttons on the bottom of the dialog, to make the
editing process easier.

Create/Edit Segment
This window is a toolbox for creating/editing a segment. To create a new segment, click the Create New button at the
top left. To edit an existing segment, drag the segment into the Drop a segment here box, or click the Browse button (
) and select it from the list. The total length of the segment and its lengths along each of the UCS axes are shown below.
The following settings are available for creating or editing the segment.

Parameters
As input, you must enter the Start point and End point of the segment. You can specify each of these points by any of
the following methods:
• Type in their coordinates manually.
• Click and then double-click on the desired point in the 3D view.
• Click and select a point from the list of already-defined points.
• Drag a point from the Project Window into the Recipe Window.
Once you have defined these points, you can enter a new Segment length. This causes the end point to extend or retract
along the segment to fit the new length.
Clicking Extract segment's three components creates three new segments that contain the x component, y component,
or z component of the original segment in the UCS. These new segments have the same name as the original segment,
but with x comp., y comp. or z comp. appended to them, as appropriate. These components are chained together, such
that the x component starts at the original segment's start point, the y component starts at the x component's end point,
and the z component starts at the y component's end point and ends at the original segment's end point.
When you are finished editing or creating a segment, simply close the Recipe Window; your changes are saved
automatically. You can also use the Save segment to menu to save it as a new segment, or to copy information about it
to your clipboard, a TXT file, or a CSV file.

Create/Edit Circle
This window is a toolbox for creating/editing a circle. To create a new circle, click the Create New button at the top left.
To edit an existing circle, drag the circle into the Drop a circle here box, or click the Browse button ( ) and select it
from the list. The length (that is, the perimeter) and area of the circle are shown below. The following settings are
available for creating or editing the segment.
Editing method
There are two ways to automatically create a circle under Editing method.

Fit circle to samples

In this method, ATLAScan creates a circle based on a large number of points that you select in the 3D view. First, click
Quick selection, then click in the 3D view and hold the mouse button down as you draw around the points that you want
to comprise the circle. When you have selected the points, click on Fit circle to create the circle. You can alter how
ATLAScan calculates the circle by clicking Fitting settings.
Fit circle to point list

In this method, you can create a circle from a list of existing points. To define these points, click Define Points to open
the Fit circle to points window, then hold down Ctrl and click on each of the desired points. Click Select points to close
the window, then click OK to fit the circle. You can alter how ATLAScan calculates the circle by clicking Fitting settings.
Parameters
This section shows the center of the circle (in UCS coordinates) and its radius. You can edit any of these values to
translate or resize the circle.
Duplicate circle as
When you are finished editing or creating a circle, simply close the Recipe Window; your changes are saved
automatically. You can also click New circle to save the current circle as a new circle.

Create/Edit Polyline
This window is a toolbox for creating/editing a polyline. To create a new polyline, click the Create New button at the top
left. To edit an existing polyline, drag the polyline into the Drop a polyline here box, or click the Browse button ( ) and
select it from the list. The number of vertices and the length of the polyline are shown below.
Editing method
There are several tools for creating and editing a polyline under Editing tools. Clicking on a tool shows its options in the
Tools Options section.

Basic tools
Add Vertices: With this tool, you create the polyline by double-clicking on each point in the 3D view that you
want to include in the polyline.
• Choose string to continue: Clicking this button then clicking on the endpoint vertex of a string in the 3D
view lets you start adding new vertices to an existing string.
• Start new string: Finishes the current string and starts a new one.
• Constrain new vertices on this plane: Projects 3D selected vertices in a plane.
• Constrain new segments orthogonally to the UCS: Forces each new segment to be orthogonal to the x,
y, or z axis of the UCS.
Edit Vertex: When this tool is selected, double-clicking on a vertex in the 3D view shows its coordinates in the
Toop Options section. You can edit the vertex by manually entering new coordinates. You can also move the
vertex by clicking and double-clicking on the new location in the 3D view. Finally, clicking lets you use
the coordinates of an existing point for the vertex.
Connect Vertices: This tool lets you select two existing endpoint vertices, which ATLAScan then joins with a
segment.
Remove Vertex: When this tool is selected, double-clicking on a vertex deletes it. If the vertex was connected
to two other vertices, those vertices are joined directly to each other with a new segment.
Remove Segment: When this tool is selected, double-clicking on a segment deletes the segment. If the
segment was connected to two other segments, those segments are not join to each other.
Grouping tools
Merge Polylines into One: When this tool is selected, double-clicking on another polyline in the 3D view
merges it with the current polyline.
Explode Polyline Strings: When this tool is selected, double-clicking on a string in the current polyline
removes all other strings in the polyline and saves them as separate polylines. The selected string stays in the
existing polyline.
Close Polyline: When this tool is selected, double-clicking on a string joins its endpoints together with a new
segment.
Grouping tools
Set Same Altitude to All Vertices: When this tool is selected, double-clicking on a polyline in the 3D view
moves its vertices to the specified Altitude along the Z axis.
Project Polyline on Plane: Clicking Project moves the vertices of the polyline onto the plane specified in the
dropdown menu along the plane's normal.

Create/Edit Sphere
This window is a toolbox for creating/editing a sphere. To create a new sphere, click the Create New button at the top
left. To edit an existing sphere, drag the sphere into the Drop a sphere here box, or click the Browse button ( ) and
select it from the list. The surface area and volume of the sphere are shown below. The following settings are available
for creating or editing the sphere.
Editing method
There are several ways to define the sphere. Selecting one of the options causes the window to display tools for creating

the sphere using that method.
Fit sphere to samples

In this method, ATLAScan creates a sphere based on a large number of points that you select in the 3D view. First, click
to be in the sphere. When you have selected the points, click on Fit sphere to create the sphere. You can alter how
ATLAScan calculates the sphere by clicking Fitting settings. If The fitted sphere must have radius checkbox is selected,
ATLAScan forces the sphere to have the specified radius.
Fit sphere to point list

In this method, you can create a sphere from a list of existing points. To define these points, click Define Points to open
the Fit sphere to points window, then hold down Ctrl and click on each of the desired points. Click Select points to close
the window, then click OK to fit the sphere. You can alter how ATLAScan calculates the circle by clicking Fitting settings.
From circle
In this method, you can create a sphere by selecting an existing circle, which is used as the sphere's equator.
Parameters
This section shows the center of the sphere (in UCS coordinates) and its radius. You can edit any of these values to
translate or resize the sphere.
Duplicate sphere as
When you are finished editing or creating a sphere, simply close the Recipe Window; your changes are saved
automatically. You can also save it as a duplicate sphere, as a mesh, or as a spherical camera. If you save it as a mesh,
you must specify the Max Approx Error for ATLAScan to use when creating the mesh.

Create/Edit Cylinder
This window is a toolbox for creating/editing a cylinder. To create a new cylinder, click the Create New button at the top

left. To edit an existing cylinder, drag the cylinder into the Drop a cylinder here box, or click the Browse button ( ) and
select it from the list. The surface area and volume of the cylinder are shown below. The following settings are available
for creating or editing the cylinder.
Editing method
There are several ways to define and edit the cylinder. Selecting one of the options causes the window to display tools
for creating the cylinder using that method.
Fit cylinder to samples

In this method, ATLAScan creates a cylinder based on a large number of points that you select in the 3D view. First, click
to be in the cylinder. When you have selected the points, click on Fit cylinder to create the cylinder. You can alter how
ATLAScan calculates the cylinder by clicking Fitting settings.
Fit cylinder to point list

In this method, you can create a cylinder from a list of existing points. To define these points, click Define Points to open
the Fit cylinder to points window, then hold down Ctrl and click on each of the desired points. Click Select points to close
the window, then click OK to fit the cylinder. You can alter how ATLAScan calculates the cylinder by clicking Fitting
settings.
Change axis
With this tool, you can edit the cylinder by selecting a premeasured distance to use as its axis. This can change the
cylinder's position, orientation and height (but its radius remains constant).
Change base circle

With this tool, you can edit the cylinder by selecting an existing circle to use as its base circle. This can change the
cylinder's position, orientation and radius (but its height remains constant).
Parameters
This section shows the radius and height of the cylinder. You can edit any of these values to alter the cylinder. Selecting
Feature extraction and clone commands opens up the Extract cylinder's features and Duplicate cylinder as sections.
Extract cylinder's features

• Extract axis: Extracts the cylinder's axis as a segment.
• Extract top circle: Extract's the cylinder's top circle as a circle object.
• Extract bottom circle: Extract's the cylinder's bottom circle (that is, its base circle) as a circle object.
Duplicate cylinder as
When you are finished editing or creating a cylinder, simply close the Recipe Window; your changes are saved
automatically. You can also save it as a duplicate cylinder, as a mesh, or as a cylindrical camera. If you save it as a mesh,
you must specify the Max Approx Error for ATLAScan to use when creating the mesh, and whether or not you want its
top and bottom circles exported as surfaces, which creates a closed cylinder.

Create/Edit Cone
This window is a toolbox for creating/editing a cone. To create a new cone, click the Create New button at the top left.
To edit an existing cone, drag the cone into the Drop a cone here box, or click the Browse button ( ) and select it from
the list. The base radius, surface area and volume of the cone are shown below. The following settings are available for
creating or editing the cone.

Editing method
There are several ways to define and edit the cone. Selecting one of the options causes the window to display tools for
creating the cone using that method.
Fit cone to samples

In this method, ATLAScan creates a cone based on a large number of points that you select in the 3D view. First, click
to be in the cone. When you have selected the points, click on Fit cone to create the cone. You can alter how ATLAScan
calculates the cone by clicking Fitting settings.
Fit cone to point list

In this method, you can create a cone from a list of existing points. To define these points, click Define Points to open
the Fit cone to points window, then hold down Ctrl and click on each of the desired points. Click Select points to close
the window, then click OK to fit the cone. You can alter how ATLAScan calculates the cone by clicking Fitting settings.
Change axis
With this tool, you can edit the cone by selecting a premeasured distance to use as its axis. This can change the cone's
position, orientation, height and radius (but its vertex angle remains constant).
Change base circle

With this tool, you can edit the cone by selecting an existing circle to use as its base circle. This can change the cone's
position, orientation, radius and vertex angle (but its height remains constant).
Parameters
This section shows the radius and vertex angle of the cone. You can edit any of these values to alter the cone. Selecting
Feature extraction and clone commands opens up the Extract cone's features and Duplicate done as sections.
Duplicate cone as
When you are finished editing or creating a cone, simply close the Recipe Window; your changes are saved
automatically. You can also save it as a duplicate cone, as a mesh, or as a perspective camera. If you save it as a mesh,
you must specify the Max Approx Error for ATLAScan to use when creating the mesh.
Extract cone's features

• Extract axis: Extracts the cone's axis as a segment.
• Extract base circle: Extract's the cone's base circle as a circle object.
Duplicate cone as
When you are finished editing or creating a cone, simply close the Recipe Window; your changes are saved
automatically. You can also save it as a duplicate cone, as a mesh, or as a perspective camera. If you save it as a mesh,
you must specify the Max Approx Error for ATLAScan to use when creating the mesh, and whether or not you want its
base circle exported as a surface, which creates a closed cone.

Distance Between Shapes

This tool measures the distance between two geometric shapes and similar objects, such as points, circles and cameras.
To measure two shapes, hold down Ctrl and click on them in the Project Window, then click Outputs > Geometric shapes
tools > Distance between shapes. Alternatively, if you click Distance between shapes without two shapes selected,
ATLAScan will open a window to let you select two shapes from a list of recently used shapes.
The measurement is made from the closest part of each shape. The measured distance is shown in a popup window, and
is stored as a new distance object.

Intersect Geometric Shapes

This tool finds the point at which two geometric shapes intersect, such as circles, segments, and cones. To intersect two
shapes, hold down Ctrl and click on them in the Project Window, then click Outputs > Geometric shapes tools >
Intersect geometric shapes. Alternatively, if you click Distance between shapes without two shapes selected, ATLAScan
will open a window to let you select two shapes from a list of recently used shapes.
If the two shapes intersect at more than one point, ATLAScan selects one of the points. The point is stored as a new
point object.

Edit Points List

This tool shows the points in the project and lets you delete and add them. To open the window, select the points or
groups you want to analyze in the Project Window and click Outputs > Geometric shapes tools > Edit points list. The Edit
Points window opens.
In this window, you can select view the name and coordinates for all of the points in each group of points. If you selected
a set of one or more individual points, they are listed in the Manually selected points group.
The following tools are available in the top bar.
Creates a new point.
Removes the selected point.
Views the selected point in the 3D view.
Imports a list of points from a TXT file.
Exports the currently points into a TXT file.

Polyline from Point List

This tool creates a polyline from a list of points. To open the window, select the points that you want to use in the
Project Window and click Outputs > Geometric shapes tools > Polyline from point list. If you did not select any points, a
window will appear with a list of points for you to choose from.
The points are joined in the same order that you select them in the Project Window. For instance, if you select the points
A through G in the order B-C-D-E-A-G-F, the segments will join the vertices in the order BC-CD-DE-EA-AG-GF.

Flythrough from Point List

This tool creates a flythrough from a list of points. To open the window, select the points that you want to use in the
Project Window and click Outputs > Geometric shapes tools > Flythrough from point list. If you did not select any
points, a window will appear with a list of points for you to choose from.
The points are joined in the same order that you select them in the Project Window. For instance, if you select the points
A through G in the order B-C-D-E-A-G-F, the segments will join the vertices in the order BC-CD-DE-EA-AG-GF.

Tunnel survey
Cylinder virtual scan This performs a virtual scan in a cylindrical projection using a cylindrical projector
as the virtual scanner.
Generate cross sections This tool generate a multiple cross section (with defined spacing between planes).
All of the planes are normal to a selected flythrough, a trajectory that describes the
elongation of a tunnel.

Video Record
ATLAScan enables you to create a video of your 3D model by selecting Outputs > Video Record:
• Flythrough Editor (shortcut key Ctrl+F): Defines and edits a trajectory that the “virtual eye” of the 3D world can go
through.
• Play: Shows a preview of your video inside the 3D window.
• Edit: Manages the key points of the flythrough and the output videos's parameters
• Make movie: Extracts the video in a viewable file.

Flythrough Window
This window allows you to define and edit a trajectory that the “virtual eye” of the 3D world can go through, generating
a flythrough video. You can input a sequence of key points that you want the virtual camera to pass through to create
impressive flythrough videos of your models. A trajectory is made up of a sequence of view points (called Keys) which
are then interpolated.
The dialog has the following features:

• Add key: Appends the current view point to the list
• Insert key before: Inserts the current view point before the selected key in the list
• Delete key: Deletes the selected key
• Clear: Clears the list
• Update trajectory: If at least 2 keys are defined, a temporary trajectory is computed for the checked keys (uncheck a
key to skip it) that can be simulated (using Play from) or added to the project (using Add to project)
• Play from: If a valid trajectory exists, an animation of the trajectory rendered, starting from the selected key, using
the desired speed.
• Stop: Stops the current simulation.
• Movie: The Movie dialog is shown to select the parameters to create a playable movie file from the flythrough.
• Add to project: Adds the current trajectory to the project.
• Close: Hides the window.
Double-click on a key to jump to its view point. If a flythrough exists in the project tree, selecting Edit in its context menu
shows the Flythrough Window with the keys that make the trajectory.
See also flythroughs and movie dialog.

Movie Dialog
This dialog allows you to create a movie out of a video trajectory in your 3D scene. Here, you can enter the video's file
name, resolution, and frames per second (FPS). Clicking Create causes a dialog to appear asking you to select a video
codec (between those present in your PC) and various compression parameters. After you select the codec and
parameters, ATLAScan begins encoding the video.
See also flythroughs, and flythrough editor.

Drawing tools
Constrain draw to plane You must select a plane in the Project Window before using this function. This
function picks 3D points in the scene in a way that their projections on the chosen
plane are selected and shown in the point list window.
AutoCAD Link AutoCAD with the Kubit plugin must be running before you activate this function.
When the link is enabled, all the points selected in Point Pick mode, in the Virtual
Scan dialog, or in the Edit 2D grid dialog are sent to AutoCAD.

Save snapshot
Save This function captures a snapshot of the 3D scene with its rendered models. You can save
the captured snapshot in the PNG, BMP, JPG, or TIFF format.
snapshot

Manage units of measures

This dialog allows you to set the preferred units of measure for lengths, areas and volumes; and to set how many
decimal digits should appear in ATLAScan's output data. This function, therefore, gives you a great amount of control on
how lengths, areas and volumes should be formatted in all ATLAScan windows, outputs, PDF reports, etc.
Presets
On the top of the above dialog, you can activate some presets:
• SI (International system of units) will set meters as length unit, square meters as area unit and cubic meters as
volume unit.
• Imperial system of units will set inches, square inches and cubic inches as length, area, and volume unit
respectively. This button features a popup menu that gives you other two options: to use feet and to use
yards as basic measure unit for length, area and volume.
• U.S. customary units will set the U.S. survey feet as predefined length unit, the U.S. square survey feet as area
unit, and the U.S. liquid gallons as volume unit.

Selecting the units of measure

In the second box of the dialog, you find three controls to select the length unit, the area unit, and the volume unit to be
used across all ATLAScan dialogs, visualized data and outputs.
You can select among a long list of measure units, belonging to the international system of units, to the British Imperial
system, and to the United States customary units, used for survey. Please see below the complete table of the units of
measure supported by ATLAScan.
Selecting the data's decimal precision

The third option in the dialog allows you to select how many decimals to visualize after the comma for each linear and
angular value output by ATLAScan. This gives you a lot of flexibility to format your output reports according to the
precision you desire, but also it also requires you to know the precision limitations of your sensor in the particular
dataset you are working on.
Define georeferencing system

Defines the type of coordinate system used in ATLAScan. This can be the local system (XYZ), WGS84 (XYZ or
longitude/latitude/altitude) or UTM (easting, northing, altitude).
Units of measure supported by ATLAScan
Length units
Name Suffix Amount in meters
Meters m 1
Centimeters cm 0.01
Millimeters mm 0.001
Kilometers km 1000
Thousandths of an inch th 0.0000254
Inches in 0.0254
Feet ft 0.3048
Yards yd 0.9144
Chains ch 20.1168
Furlongs fur 201.168
U.S. links li 0.2012
U.S. survey feet US ft 0.30480061
U.S. rods rd 5.02921
U.S. chains US ch 20.11684
U.S. furlongs US fur 201.1684

Area units
Name Suffix Amount in square meters
Square meters m² 1
Square centimeters cm² 0.0001
Hectares ha 10000
Square kilometers km² 1000000
Square inches in² 0.00064516
Square feet ft² 0.09290304
Square yards yd² 0.83612736
Acres acres 4046.8564224
U.S. square survey feet US ft² 0.09290341
U.S. acres US acres 4046.873
Volume units
Name Suffix Amount in cubic meters
Cubic meters m³ 1
Cubic centimeters cm³ 0.000001
Cubic millimeters mm³ 0.000000001
Liters l 0.001
Cubic inches in³ 0.000016387064
Cubic feet ft³ 0.028316846592
Cubic yards yd³ 0.764554857984
Gallons gal 0.00454609
U.S. cubic feet US ft³ 0.02831685
U.S cubic yards US yd³ 0.764554857984
U.S. acre feet US acre feet 1233.482
U.S. liquid gallons US fl gal 0.003785411784
U.S. liquid barrels US fl bbl 0.119240471196
U.S. dry gallons US dry gal 0.004404884

Tools menu
The Tools menu contains several processing tools. You can find most used tools also in the Top Toolbar.
• Go To Process
o File
o Navigation
o Tools
• Pose & Registration
• Selection & Fitting Tools
• Points Filtering & Clustering
• Meshing
• Photo & Color

Process Environment
The Process environment is a tool in ATLAScan that lets you do the following to any set of points clouds (both
unstructured and structured grid point clouds):
• Import
• Pre-process
• Register
• Georeference.
In general, your workflow is assisted by following the steps illustrated here:
You can enter the Process environment by clicking Tools > Go to Process.
The Process menu is subdivided into the following command groups:

• File
• Navigation
• Tools

In the Top toolbar you can find the most used commands, grouped by main processes:
Launcher: Closes this project and goes back to the Launcher.
Control: Goes to the Control screen.
Process: Opens the Process environment.
Extract: Goes to the main ATLAScan environment.
Save Project: Saves the project.
Parse: Opens the Parse window.
Camera: Opens the Panorama creator window.
Wizard: Opens the Scan Processing Wizard.
Register tools: Registers the imported scans. This command includes both pre-registration process
(target-free or target-based) and fine registration
Georeference tools: Georeferences the models using known control points
Measures: Takes measurements on the point clouds
Predefined Views: Switches to a predefined view.
Screen Settings: Selects the settings for the 3D view.
View Parameters: Selects the projection and cut planes for the 3D view.
Selection tools: Select a region of a point cloud and deletes the area inside or outside it

File menu in Process

Open current project folder: Opens the project's folder in Windows Explorer.
Save Project: This function saves the current project, safely storing all the project items, their
load/unload state, their properties, the current 3D camera pose, etc. It is advised to save the project
frequently, especially before long processing operations.
Import point clouds: Imports a point cloud in one of the formats supported by ATLAScan.

Scans Processing Wizard

This wizard allows you to import, preprocess, preregister and register any set of scans.

Navigation in Process
Navigation in the Process environment tool is very similar to that in the
regular ATLAScan environment:
• Predefined views
• Navigation modes
• Screen settings
In addition, other predefined views are available:

• Slice View
• Horizontal Section View
These modes can be used to check the correct alignment along a

vertical/horizontal section of the displayed items.
The Manage units of measure tools is also present.

Tools in Process Environment

The Process environment features the following categories of tools.
Point Filtering & Clustering

Filters/edits point clouds and clusters them.
Pose & Registration

Registers and georeferences the point clouds and 3D models.
Select
Selects a region of point cloud and deletes the area inside or outside it.
Measures & notes

Measures distances and takes screenshots of the 3D window.
Settings
Change all the parameters of the different steps of the workflow (see “More Settings” in the Scan
Processing Wizard).
Wizard
Opens the Scan Processing Wizard and imports, preprocesses, preregisteres and registers any set
of scans.
See also the items in the general Tools command in the Extract environment.

Points filtering and clustering in the Processing

environment
This category of tools includes functions dedicated to working with point clouds to enable other processes and further
results. Most of these tools work on any point cloud, but some of them work only on grid point clouds. In the Processing
environment the following filtering and clustering tools are present:
• Pre-process grids
• Restore raw data
• Restore deleted points
• Point clustering
• Make single cloud
• Remove duplicate points

Pose & Registration in Process Environment

The Process Environment includes a comprehensive suite of tools for coarse and fine registration of any number of scans
(and not only). The registration process is composed by several steps, depending on adopted survey techniques.
The general flow, illustrated below, starts from a preregistration step that allows you to compute a rough alignment
between the imported models. The alignment can be later refined using ICP registration and Bundle Adjustment to
choose and refine the good ICPs and discard the wrong ones, in order to reduce the global registration error.
Registration
• Automatic, target-free preregistration
• Targets registration
• Manual pre-registration
• Manual pre-registration among grid point clouds
• Cloud-to-cloud registration
• Bundle Adjustment (Global fine registration)
• Scan alignment per groups
Georeferencing
• Point cloud georeferencing
• Manage control points
Positioning
• Pose transform
• Manage circular targets
• Determine the cloud's vertical direction
• Manual positioning (adjust Pose)

Automatic Target-free Pre-registration

This dialog enables a very effective and fast procedure to preregister a set of scans automatically, without using targets
or markers.

Features
• No limits on the number of scans to preregister. You can preregister 200 scans at the same time; the software will
load at most two of them at a time.
• Easy convergence with automatic search of best overlap. To be registered together, the minimum overlap of two
scans is 20-30% of their surface. However, you don't need to worry about this in most cases: given a set of N scans,
the algorithm is guaranteed to find all the scan pairs that match together and to use the best overlapping N-1 scan
pairs to preregister your dataset. You can customize this behavior via the Maximum distance in scan sequence
option.
• No point density requirements. The algorithm adapts to the specific datasets, whether they contain close-range
indoor data or long-range outdoor data.
• No complicated parameters to input such as cell size or fitting strategy. You only need to specify whether your
dataset contains "construction" or "territorial" data: the algorithm will change accordingly the surface elements to
look for and to match.
• Very fast because of parallel computing: The algorithm is guaranteed to do all computations in parallel, using all the
cores your CPU has (unless you assign fewer cores to ATLAScan using Windows).
• Robust against symmetric datasets. In case of rectangular empty rooms, stairs, or other symmetric datasets,
ATLAScan guaranteed to quickly let you browse multiple preregistration solutions and help you quickly find the right
preregistration of your scans.
• Easy to set constraints. You can easily set constraints on the scan vertical. If you know that the scans are already on
their vertical (with a confidence of 2-3°), you can tell the algorithm to leave the scans' vertical unchanged. If you
don't, ATLAScan will also estimate the scans' vertical, with the guarantee to correct angles between the given
vertical and the correct one of max. 30°. As additional constraint, you can select which of the input scans is the
reference scan during the preregistration.
• It works for virtually all datasets. This automatic preregistration technique has been tested with hundreds of
different datasets containing from 10 to 200 scans per dataset. It gets your scans in a state of coarse registration
without your intervention, or with very few easy corrections at the end.
Parameters
Here we give a top-to-bottom description of the parameters that you can set in the dialog.
• Data Type: here you can specify if your datasets relates more to a architectural scene with many planar surfaces, or
if it relates more to a mining/terrain scene with many irregular surfaces. This will greatly help the algorithm.
• Speed vs precision: This algorithm is designed to be fast in the first place, and to deliver a solution in minutes or
tens of minutes. Generally, you should leave this option as Fastest and to switch to the other two options only when
the first does not work. The other two options make the algorithm search for more matches between smaller
surface elements. This leads in general to better results, but in some cases it can also spoil the results because it
introduces more candidate matches for the final solution and therefore more ambiguities.

• Preregistration options:
o Default preregistration: With this option, ATLAScan to perform preregistration without a preliminary
sketch or constrained origins
o Preregistration from sketch: With this option, you can help the preregistration algorithm by providing a
preliminary “sketch” that shows the approximate relative positions of the scans. To create or load a sketch,
press the Create/Load sketch button. The Sketch Window opens for you to create the sketch.
o Preregistration with constrained origins: With this option, you can help the preregistration algorithm by
providing origins for the scans with certain restraints.
• Assume vertical is correct: If your Polaris is using its integrated inclinometer, it can create raw scan data where the
vertical is already correct, with an error of 2-3 degrees. If this is the case, you can set this option to make the
algorithm match the scans without changing their initial vertical. If the inclinometer was not used, the the algorithm
will also estimate the correct vertical before matching the scans. Please take into account that the maximum vertical
error that can be corrected is 30 to 40 degrees.
• Show more options: This opens a small window with the following advanced options.
• Save poses before preregistration: If you flag this option, ATLAScan will automatically save the scans' poses in the
project folder's Export subfolder before starting the automatic preregistration. In this way, you will be able to easily
go back to those poses before the preregistration by selecting Tools > Pose and registration > Pose transform >
Apply pose from file.
• All scans are tripod (no hand-held or mobile): Flag this command if you're only using tripod laser scans.
• Maximum distance in scan sequence: This parameter determines how many scan pairs the algorithm will try to
match. It specifies the maximum distance in the scan sequence that two matchable scans may have, and it defaults
to 10. For example: given an input sequence of 30 scans, ATLAScan will try to match scan 5 with all of the following
scans up to scan 15. Scan 5 and scan 16 will not be matched because ATLAScan will assume that they are too far
away in 3D, being far away in the sequence. Generally speaking, given an input sequence of N scans and this
parameter set to M, then ATLAScan will try to match scan s only with the M scans preceding and the M scans
following it in the input sequence, considered as a circular sequence (scanN matches scan 1).
• Scan sequence is closed: Designates the scan sequence as being closed.
• Rotate all scans on their verticals when estimating verticals

Sketch Window
This dialog enables a very effective and fast procedure that preregisters a set of scans automatically, without using
targets or markers.
This window enables creation, editing and handling of preregistration sketches, created in ATLAScan. Preregistration
sketches are needed to describe the approximate relative positions of a set of scans, to give an initial hint to ATLAScan's
automatic target-free preregistration. However, it is not compulsory to create a preregistration sketch to run the
automatic target-free preregistration.
Usefulness
Creating or loading a preregistration sketch is useful mainly to speed up the automatic target-free preregistration : in
fact, if the scans' approximate relative positions are known, the automatic preregistration algorithm will match only the
scan pairs that belong to the registration graph: the blue polyline connecting the scans in the upper picture. Therefore,
to preregister N scans the algorithm will match only N – 1 scan pairs instead of trying all the combinations, which can
result in N ( N – 1) matches. Therefore, automatic preregistration will be faster and will aim at the right solution more
directly, avoiding matching the wrong scans with the subsequent ambiguity and confusion.
Another usefulness lies in keeping a sketch of your scans' positions. Yet another advantage of this window lies in the
possibility of setting which scan is the reference scan, and optionally which scans should not be used during the
registration.

How it works
In the main area of the sketch window, the sketch is shown. The scans are represented as icons on the sketch related to
the scanner model. Below each icon, the scan name and the scan registration role are found. Each scan can have one of
three roles:
• Reference scan. The scan is not moved during the automatic preregistration.
• Moving scan. The scan will be preregistered against the others during the preregistration.
• Scan not used. The scan will not be preregistered. This is useful if you acquired scans that are not useful for
the preregistration, like hi-resolution scans of small targets, or test scans.
Simply drag the scan icons around the sketch to change their position. You can change the scans' registration roles via
the Edit registration roles button on the left.
On the right, some buttons are present.

• Open sketch, Save sketch, Save sketch as. These buttons manage your sketch files, which have the SKETCH
suffix. You can create a preregistration sketch from scratch and then save it.
• Show registration graph. Toggle this button to show or hide the blue polyline that represents the best
registration open sequence for your scans. Move the scans so that the registration graph is the best possible.
• Sketch is OK. Select this option to close the sketch window and to go back to the control dialog for starting
the automatic preregistration taking as input the edited sketch.
• Cancel. Select this option to close the sketch window and go back to the automatic preregistration dialog
without using the sketch.

Edit Registration Constraints Window
This dialog enables you to provide a list of origins and constraints for the scans. This is optional, but it improves the
performance of ATLAScan's automatic target-free preregistration algorithm.

Targets registration
This function automatically registers a set of scans by detecting and matching targets. You can start the automatic target
detection, or manually add, move and remove targets.
For each selected scan, the user can see the number of detected targets. By clicking the eye icon, the user can open the
Target editor dialog for the selected scan.
The Detect Targets button allows the user to perform the target finding stage on the selected scans. The targets
associated with the selected scan are overwritten.
The Register Scans button performs both the targets detection and scan registration stages. The detection phase is
skipped for scans with already associated targets.
Options
• Detect circular targets: This option makes ATLAScan attempt to locate circular targets in the point cloud
automatically. If this is selected, you can enter the diameters of the targets below.
• Targets diameters selection: The user can add one or more diameter values by typing the numeric value and
pressing the button. An inserted diameter can be removed by selecting it from the list and clicking the button.
• Targets matching: Matches the targets either automatically or based upon their names.
• Alignment options: The user can select whether the scan's vertical is to be assumed known.
Note: It is necessary before to compute the confidence of the point clouds (see Pre-processing Grids) to help the process
algorithm to find the target. At least 5 point cloud's points (on the same polyline) are needed to fit the contour of the
target (an ellipse).

Target Editor
The targets editor dialog shows a 2D view of the current scan with its associated target points. You can manually add,
move and remove targets. If you enter this window through the Manage circular targets command, the Find targets
button is also displayed, which lets you locate the targets automatically.
The mouse controls are the following:

• Double-click to add a target point
• Right-click to remove an existing target
• Drag the cursor while pressing the left mouse button to move an existing target.
On the right side of the window are listed the current targets with their 3D coordinates and confidence values. The zoom
window can be centered on a specific target by clicking on a target from the list. A target's label can be edited by double
clicking the respective cell in the targets list. Note that a target label must be unique among the targets in the same
point cloud.
It's also possible to export points (in either UCS or local coordinates) and import points with buttons at the top.

Targets Registration Report
This dialog allows you to refine and exploit the result of a target-based registration. On top of the dialog, the mean
registration error is shown, along with the editable error threshold.
In the central table of the dialog, the matched targets are listed, grouped by scan. Note that each target can have one or
more match with targets from other point clouds.
You can uncheck a matching target pair, by pressing the checkbox on the right, and press Update registration on the
bottom left. The registration will be recomputed without the outlier and the mean registration error will improve
accordingly.
Please note that the alignment between clouds might be lost when matches are unchecked. In any case you can restore
the original computed registration by clicking the Reset To Default button in the bottom part of the dialog.
Once you are satisfied with the computed alignment, you can apply the registration and choose to save a PDF report.
You can also uncheck multiple matches at the same time by selecting an error threshold and clicking the Remove Above
Threshold button in the top right corner of the dialog.

Manual Pre-registration
Manual preregistration quick guide

1. Select at least two models from ATLAScan project view.
2. Select “Manual preregistration” among the registration tools. The dialog above appears.
3. In the list on the bottom right, select the reference model and the moving model. These two models are
shown in the respective 3D views.
4. Navigate as usual in the 3D views, and
5. Find at least three pairs of corresponding points between the reference and the moving models.
Double-click with the left mouse button on a model to select a point.
Right-click with the mouse on a point to remove it.
6. Press the Compute button on bottom right, and select Apply transform if the registration error is satisfactory.
The registration error doesn't need to be good, because it can be refined via the ICP registration.
Introduction
This function is part of ATLAScan's registration techniques. This pre-registration technique allows you to manually
compute a rough alignment between two models (unstructured point clouds, structured point clouds, or triangle
meshes). The alignment can be later refined automatically, using ICP registration. This function is useful in the rare cases
when the Automatic preregistration function fails.

Reference and moving models

To use this function, please select from the project view at least two models. The models can be of any type: point
clouds (structured or not) and/or triangle meshes.
When the preregistration dialog opens up, the selected models appear in the Models to register list on the bottom right.
Please select a model as reference and a model to be preregistered against the reference. When you select a model
either as reference or as moving, the model is rendered in the corresponding 3D window. Please use the buttons above
each 3D window to switch between a default view and a bubble view of the model.
You can change the reference and moving models at any time.
Picking pairs of homologous points

To preregister the models, you indicate to ATLAScan features that the models have in common, such as building edges,
window corners, etc. Double-click with the mouse on a model to select a specific feature point. The point gets displayed
in the 3D view with its label, and gets listed in one of the point tables on the bottom of the dialog, with its label and its
Cartesian coordinates shown in the current unit of measure. Right-click with the mouse on a point in the 3D scene to
remove it from the list.
When you have at least three point correspondences, the error table on the bottom of the window displays the
registration error associated with each point, to offer you a way to remove the worst point pair and select another one
to improve the registration.
Finishing the preregistration

When at least 3 point correspondences are found press Compute. A dialog shows you the mean registration error and
gives you the options to immediately apply the registration to the moving model, to copy the registration transform to
clipboard to apply it later, or to cancel if you don't accept the error. You may want to copy the transform to the
clipboard if you want to move more models according to the same transform. This is useful for example if you have a
group of say 10 grids already registered among them, to be registered against other 10 already registered among them.
To apply “later” a registration transform stored in the clipboard, select in the project window the grids you want to apply
the transform to, right-click on them and select Registration->matrix transform from the context menu. Then
select Apply registration transform in the Matrix transform dialog.
When the pre-registration is finished, the models are unloaded from the project to free memory. Please reload them to
see them aligned.
Registration constraints
On the bottom right of the dialog, you can also add some registration constraints. You can specify whether or not the
vertical of the moving model is fixed. If you do so, you only need to specify at least two point correspondences instead of
three to compute the preregistration. Additionally, you can specify the precision of the moving model origin's position,
among four levels. If you specify that the moving model's origin is fixed, then you only need one pair of corresponding
points to get a preregistration.
Computing mutual rotation among models

There are cases in which two models just need to be rotated around the vertical by some angle each, to get into a state
of coarse registration. Many laser scanners embark GPS and inclinometers, therefore in some situation you may already
possess the approximate values of the scans' vertical and origin. If that is the case, you can set the following registration
constrains: fixed vertical, fixed center and de-flag the option reference orientation is correct. If you do so, you only need
to select at least one pair of corresponding points, and ATLAScan will compute the rotation angles around the vertical for
both the reference and the moving model (the reference model will be moved as well).

Manual Pre-registration (among grid point clouds)
The pre-registration technique allows you to manually compute a rough alignment between two grid point clouds. The
alignment can be later refined automatically, using ICP registration.
Defining the reference and the moving grid

From the list of grids in the dialog, choose the reference model and press Set reference grid, then select the grid to
move and press Set moving grid. The two grids appear as range images on the right and left panel of the dialog
respectively.
Finding corresponding points

The preregistration procedure works by finding three couples of corresponding points among the reference and moving
grids. If needed, you can register one moving grid to more reference grids: to do so, change the reference grid to
another from the loaded ones, the image-grid point pairs are stored for each grid, and are accumulated.

Keyboard/mouse event Effect
LMB double click Selects a point. Selected points are listed in the
tables moving grid and reference grid at bottom of the
dialog
Mouse wheel rotates Zooms the range image in/out centering on the hovered
point
Key 'z' Toggles the Zoom views shown in the screenshot
Ctrl + LMB + move mouse Moves the zoomed region inside the global view
Space bar Rotates the range image of 90°
Shift + LMB + move mouse Translates the range image
RMB on a selected point Deletes the point
LMB on a point + move mouse Moves the point around
Alt + central mouse button Fit image to window
Alt + RMB Reset image to original size
When three or more couples of points are selected, the table Error per point shows the alignment error per point
couple. The mean registration error is instead shown at bottom left of the dialog. In general a pre-registration error of 2
cm is acceptable. However, this depends on the scale of your project and of the final output you want to get. If a point
couple introduces too much error, you may try to move slightly around the points after zooming on them, to improve
that couple's error.
Try to change the color type of the grid on top of the window to improve the contrast of features. Press the right arrow
to show a menu of available commands for the current color type for better adjustment.
The list of points is shown in the table, where every cell is manually editable to force the values, like for geo-
referentiation.
Finishing the preregistration

When at least 3 point correspondences are found, press Compute. A dialog shows you the mean registration error and
gives you the options to immediately apply the registration to the moving grid, to copy the registration transform to
clipboard to apply it later, or to cancel if you don't accept the error. You may want to copy the transform to the
clipboard if you want to move more grids according to the same transform. This is useful for example if you have a group
of say 10 grids already registered among them, to be registered against other 10 already registered among them.
To apply a registration transform stored in the clipboard later, select the grids you want to apply the transform to in the
Project Window, right-click on them, and select Registration > Matrix transform from the context menu. Finally,
select Apply registration transform in the Matrix transform dialog.
When the pre-registration is finished, the grids are unloaded from the project to free memory. You must reload them to
see them rendered.

ICP Cloud to Cloud Registration

ICP Registration is an algorithm to automatically perform fine registration of a moving point cloud against one or more
reference clouds. The moving cloud must be roughly close to the reference cloud. It is possible to roughly align the
moving cloud by pre-registering it.
Accessing the ICP Registration dialog

You can access this dialog by selecting Tools > Pose & Registration > Cloud to cloud registration from the menu in the
Process environment.
The ICP algorithm works on both structured and unstructured point clouds. The requirement is that the point clouds
have the normals calculated, i.e. have the Inclination among their colors. If a point cloud doesn't have the inclination, ICP
will issue an error message during processing . Inclination can be computed for grid point clouds during preprocessing.
To register point clouds, just drag and drop them from the project window to the ICP dialog. The clouds will appear in
the top-left list in the dialog.
Setting the clouds' registration roles

In the top-left list, each cloud is listed with its name and its Registration role. By clicking the button at the right of any
cloud's name, you can select one of three registration roles: each cloud can be a Reference scan, a Moving scan, or
simply not used.

A reference scan will remain still in its position during registration, while a moving scan will be moved to align it better to
the reference clouds. You can have as many reference scans as you want, but you must have exactly one moving scan to
run ICP.
Example: Suppose you need to register 10 clouds together. You may want to drag and drop them all onto the ICP dialog,
then register the first two, then register the third using the first two as reference, and so on, gradually registering the
point clouds on the ones already registered.
Using a preregistration report to finely register a set of clouds

If you have a set of clouds already preregistered using automatic preregistration, you can automatically run in batch a
sequence of ICPs that will register your clouds. Press the button Start from preregistration report, then select the
preregistration report you want to use (usually found in the Results/Reports project folder). Below the button, a
confirmation message appears ensuring that the report has been correctly loaded and that you can press Process to
register all your clouds. After that you have loaded the preregistration report, you are not allowed anymore to change
the registration roles, since the reference and moving scans are already defined by the report. This technique will use
the minimum number of pairwise ICPs to register your clouds: if you have N clouds, N – 1 ICPs will be needed, as many as
the edges of the minimum spanning tree that connects all your clouds. The information about which is the minimum
spanning tree is contained in the preregistration report.
Comparison with Bundle Adjustment. ATLAScan offers two algorithms to finely register a set of point cloud starting
from a preregistration report: the one usable via this dialog, and Bundle Adjustment. Which shoud you use?
As noted, the algorithm available here uses the minimum possible number of ICPs to connect your clouds to each other.
Therefore, if your clouds form closed sequence (e.g. a loop around a building, or inside a hall), you have no guarantees
that the first cloud will be aligned with the last. Bundle Adjustment, instead, is designed to use as many ICPs as possible
to connect each scan with all its neighbours in order to globally register all your data.
In general, if your scans are positioned in a strict sequence, you should use the ICP method, because bundle adjustment
may try to register together clouds that have no significant overlap with each other. If, on the other hand, your scans
form a loop or a closed sequence, you should definitely avoid the present method and use a bundle adjustment.
Setting the ICP parameters

Click Set ICP parameters to open the ICP Parameters window. This window also allows you to save the parameters in a
separate file, which you can then load for use in later projects.When the parameters are set, click Start. ICP begins,
showing log messages in the right-central Result panel of the dialog. Also, the graph on bottom right displays the error
histogram for each iteration. This is useful to see at a glance whether the algorithm is converging.
Finishing the registration

After finishing the registration, the dialog pictured above shows you the mean registration error and offers you three
options: to immediately apply the registration to the moving grid, to copy the registration transform to clipboard to
apply it later, or to cancel if you don't accept the error. You may want to copy the transform to the clipboard if you want
to move more grids according to the same transform. This is useful for example if you have a group of say 10 grids
already registered among them, to be registered against other 10 already registered among them.
To apply a registration transform stored in the clipboard, select in the project window the grids you want to apply the
transform to, right-click on them and select Registration > Matrix transform from the context menu. Then select Apply
registration transform in the Matrix transform dialog.
When the registration is finished, the grids are unloaded from the project to free memory. You must reload them to see
them rendered.

ICP Parameters
This window enables you to adjust the parameters used during ICP cloud-to-cloud registration and Bundle Adjustment.
You can also save the parameters and load them for use in later projects.
Setting the ICP parameters

ICP stands for Iterative Closest Point. The algorithm finds points on the moving cloud that are close to the reference
clouds. These points are called control points or inliers. The algorithm then iteratively moves the moving cloud to reduce
the distance of the control points to the reference models. After each step of movement, the control points are
recomputed.
Once the reference and moving clouds are set, the user can also tune several parameters of the ICP algorithm, displayed
in the Parameters box on top right of the dialog. The default parameters should work fine for most cases. However, here
a full explanation of the ICP parameters is provided.

ICP parameters are divided in three groups: parameters that define how to search for inliers, parameters that define the
convergence criterion, and parameters that define registration constraints.
Inliers search parameters

• Min. search distance. See explanation for Number of control points.
• Max. search distance. If the models have a bad pre-registration, you can increase this value to compensate,
but note that this will slow down the algorithm.
• Number of control points. Tune the number of control points to the size of the models. The model to register
(not the reference) is sampled with Number of control points. At the beginning of ICP, a search distance S is
computed. Control points from the moving cloud are considered only if they are closer to the reference
models than S. S is calculated so that it is between Minimum search distance and Maximum search distance,
and so that at least Sufficient control points can be found in the moving cloud to be closer than S to the
reference clouds. If such an S cannot be found, for example because the clouds are too far away and
therefore the control points are too few, ICP stops and an error message is issued.
• Min. control points. During ICP iterations, the control points are recalculated. If at a certain iteration they are
fewer than this value, ICP stops and an error message is issued.
• Sufficient control points. See explanation for Number of control points.
• Max. normal divergence [deg]. Two points close to each other are considered not matching if their normals
are diverging more than this angle. This is useful to filter out from the registration noisy data like vegetation,
foliage, or moving objects like cars or people. The assumption is that points belonging to the same object
should have also the same normal direction.
• Discard normals' verses
Convergence criterion parameters

• Acceptable registration error [m]. Tune this parameter to select the desired registration accuracy. ICP will
stop iterating if the mean registration error goes below this value. The other parameters that influence
ICP's stop criterion are: Min. error change to continue and Max. iterations.
• Min. error change to continue. ICP stops if the error-delta between three iterations is smaller than this
parameter. Let's call this parameter d. Let's also define e(n) to be the mean registration error at iteration n.
ICP will stop iterating if, at iteration n, [(e(n – 3) – e(n)) / e(n) ] < d.
•
• Max. iterations. ICP will do at most this many iterations.
ICP constraints
• Vertical is fixed. Set it to true if you know that the moving scan is already set on its vertical in the current
UCS.
• The next options determine whether the algorithm is allowed to translate the origin of the moving point
cloud, and how much:
o Origin is not constrained: The algorithm will translate the origin of the cloud freely.
o Origin is known: The algorithm will only translate the origin of the cloud within the given
horizontal and vertical constraints.
o Origin is fixed: The algorithm will not translate the origin of the cloud at all. This is useful if you
know with great precision (e.g. via total station) the coordinates of the scan's origin.
Saving and loading parameters

Clicking Save params stores the parameters in this window in an XML file. If you want to use the same parameters in a
later project, you can load them by clicking Load params.

Bundle Adjustment
Solving a puzzle of a thousand pieces

To understand bundle adjustment, we can imagine to solve a big puzzle with many pieces. Each piece matches with four
others, sometimes the match is easy, sometimes you have to slightly push a piece inside another, and sometimes two
pieces just don't match. Often, you need to know the neighboring pieces to decide whether two pieces match correctly
or not. And the pieces reach their final, adjusted positions only when the puzzle is complete. If ICP registration can be
seen as finding the match between two pieces, Bundle Adjustment means completing the whole puzzle by choosing and
refining the good ICPs and discarding the wrong ones.
This algorithm allows you to register together many point clouds, evenly distributing the registration error. The user
specifies which point clouds are reference clouds (these are locked during the registration) and which are moving (these
will move and align on the reference clouds and between them during alignment). While the ICP registration works only
for one pair of clouds at a time, this algorithm registers together N clouds at the same time, diffusing and minimizing the
global registration error. The input point clouds must be pre-registered. They do not need to be structured but they need
to have the normals. There is no upper limit to the number of the input clouds.

How it works
Given a set of N scans, the algorithm will create as many connections among pairs of neighboring scans as possible. For
each connection, a pairwise ICP will be executed, using the Settings accessible from the Process environment's top
toolbar. From each ICP, the best matching point pairs between the two scans will be saved. In the end, a final nonlinear
minimization step will be run only among these matching point pairs of all the connections. The global registration error
among these point pairs is minimized, having as unknown variables the scans' poses.
Parameters
Input point clouds

Just drag the poing clouds you want to finely register from the project view, and drop them on the Bundle Adjustment
dialog. They will appear in the top list. On the right of each cloud, a button appears that can be toggled as either red
(reference scan) and green (moving scan). This lets you identify which scans must remain still during registration and
which must be registered.
Preregistration report
If you have previously performed the automatic preregistration of your scans, you can use the results of your
preregistration in the bundle adjustment. Click Start from preregistration report and load the report of the automatic
preregistration, a text file saved by default in the Exports folder of your project. The dialog will show in the top list all the
clouds that were automatically preregistered, and will adjust the ICP settings accordingly to the preregistration error
contained in the report.
Defining connections among scans

To reach the best global registration possible, this algorithm will use as many connections as possible among your scans.
Knowing the topology of your scans, you can set the Max connections per scan parameter accordingly. If you scanned
the exterior of a building, then it may be that each scan shares geometry with 3 or 4 neighbors. If, instead, you scanned a
room or a hall, then it makes sense to say that each scan matches with other 5 or 6 scans, therefore all pairwise ICPs
between a given scan and its 5 or 6 neighbors should be launched. However, in doing so there is the risk of running ICP
between two scans that have no geometry in common. In this case, the ICP error will be very high.Therefore, the
algorithm will discard bad ICPs and keep the good ones. You can set the Severity in discarding bad ICPs among three
levels as in the dialog. It is generally best to start with Strict, and only select Coarse if your data are particularly noisy
(e.g. mining data).
Global constraints
You can specify some constraints that must remain valid for all the scans during all the registration by clicking Set
constraints and ICP parameters for all scans. This opens up the ICP Parameters window.
Scan-specific constraints
The top of the window shows the list of scans to be registered. On the right of each scan, a button allows you to switch
between whether the scan should be reference or moving.
In addition to the global constraints, you can can define specific constraints for each moving scan by clicking Edit
constraints. This creates a window where you can identify whether the scan has the vertical fixed or not. Moreover, you
can constrain the scan's origin in three levels. The origin can be free to move (not constrained), can be constrained
within certain horizontal and vertical bounds, or can be held fixed.

Output
Once you have defined the connections among the scans and the additional constraints, press Start to begin processing.
A progress bar appears, and you can also open the log window to have more detail on what's happening.
At the end, the clouds will be moved to the new positions and their files will be saved. Therefore, you may want to save
the clouds' poses before running bundle adjustment. Moreover, a dialog appears indicating the number of registered
clouds and the global registration error. After you press OK, you can click on Save PDF report to get a much more
detailed PDF report. You can also save the following items in that button's drop-down menu:
• Save .txt report: Saves a text version of the full report that can be copied to Word or Excel
• Save XML report: Saves the details of the report in an XML file.
• Save registration graph: Saves the registration graph as a PNG image.
Resolution of common problems

I have run bundle adjustment, but scan A didn't match well with scan B
Get the report of the bundle adjustment, and check whether the edge between A and B (or B and A) is listed in the
section Report by edge, sorted by error. If it is, most likely the ICP between A and B has converged badly: you can re-run
the ICP between A and B only with different parameters, and then re-launch the bundle adjustment with A and B among
the references.
However, most likely the edge between A and B will be in the Edges not used section of the report, or even not in the
report at all. If the edge (A, B) is among the not used, then try to re-launch the bundle adjustment, decreasing Severity in
discarding bad ICPs. If the edge (A, B) is not in the report, then try to increase Max connections per scan and re-launch
the bundle adjustment.
I have run bundle adjustment, but I see that the scans did not converge well. Moreover the global registration error is
4 cm while it should go down to 2 cm.
Press the Settings button of the Processing environment's top toolbar, and go to the ICP settings. Set the ICP parameters
so that only the very best inliers are selected. You can lower the min search distance to 0.2 m or to 0.1 m, lower the
sufficient control points to 3000, and lower the max divergence among normals to 10° or even down to 3°. Then, re-
launch the bundle adjustment. Only the best matching points among scans will be selected, and the error among them
will be minimized, resulting in a better convergence and tighter error. You can repeat the process to improve the results.

Bundle Adjustment Report

At the end of the bundle adjustment process you can save a PDF, TXT or XML report detailing the ICP and Bundle
Adjustment results. The following information is shown in the report.
Summary
• [N] scans have been finely registered with bundle adjustment.
• Num ICPs used: Number of ICPs actually used
• Average ICP error [m]: Average value of registration errors related to all scan pairs
• Average registration error of corresponding points [m]: Global registration error reached between the most
important corresponding points. This error is the main result of the bundle adjustment.
Input parameters
In this section the Inliers Search Parameters that you set before the bundle adjustment are summarized:
• Max neighbors for each scan [N]: Number of scans (close to the current scan) taken into account
• ICP selection severity: Level of severity used in discarding bad ICPs.
• ICP min search distance [m]: See Inliers Search Parameters in ICP Parameters.
• ICP max divergence among point normals [deg]: See Inliers Search Parameters in ICP Parameters.
• List of registered clouds: The well-registered point clouds are listed here.
Detailed results
• ICPs executed: Number of ICP processes that have been executed during bundle adjustment.
• ICPs used: Number of ICP registrations that have been found good enough to contribute to the final bundle
adjustment.
• Average ICP error: Average value of registration errors related to all pairs of scans.
• Maximum ICP error allowed [m]: See Inliers Search Parameters in ICP Parameters.
• Total cost before minimization [m]: Statistical parameter derived from least squares method, relative to state
before minimization.
• Total cost after minimization [m]: Statistical parameter derived from least squares method, related to state after
minimization.
• If Total cost after minimization < Total cost before minimization an error decrease is achieved
• Reason of convergence: If the value of |cost_change|/cost is lower than 1.000000e-008 (fixed upper limit) the
function tolerance is reached.
• Average registration error of corresponding points [m]: Global registration error reached between the most
important corresponding points. This error is the main result of the bundle adjustment.
• N° corresponding points: The most important corresponding points among all valid scan pairs. The final global error
minimization is run exactly on all these corresponding points.
• N° iterations: Number of iterations in the iterative decreasing error process
• N° parameters, N° effective parameters: These are the parameters of the error functional to minimize, and depend
on the degrees of freedom of the problem. If bundle adjustment is run among ten scans for example, keeping one as
reference, then the effective parameters are 9 moving scans * 6 degrees of freedom (three for translation and three
for rotation) = 54. However, if you change the registration constraints (e.g. adding more reference scans, imposing
fixed vertical and/or fixed translation) the number of effective parameters changes accordingly.
• N° threads used: Bundle adjustment uses as many threads as the logical CUP cores of your PC. Therefore, bundle
adjustment is very CPU-greedy to offer you the fastest processing time. If you desire another behavior, you can

assign fewer processors to ATLAScan from the Windows task manager.

• Time taken for final bundle adjustment [mm:ss.msec]
• Total elapsed time [mm:ss.msec]
Report by edge, sorted by error

All the analyzed couples of scans (i.e. edge A and B) are sorted regarding the registration errors at the end of ICP and
bundle adjustment phases.
• [N] corresponding points found between scans A and B
• ICP error [m]: Registration error related to single pair of scans, as they are isolated from others.
• Correction after bundle adjustment [m]: By how much the registration error related to the single scan pair has
changed after the global error minimization.
Edges not used

If all scans are finely registered, All edges have been used is written here. Otherwise, a list of discarded edges, sorted by
error, is shown. If you see among the “edges not used” a scan pair that you think should be registered together, then you
may want to re-run Bundle Adjustment with coarser parameters. In the same way, if a scan pair has been registered
together that you think it shouldn't have, you can re-launch Bundle Adjustment with stricter parameters.

Scan alignment per groups

What you can do if the automatic pre-registration fails?
In some cases, a set of scans (and/or 3D models) cannot be aligned due to low overlap between some scans. If it is
possible to detect groups of scans already registered and one scan per group that have good overlap each other, a
manual pre-registration can be run.
A group is a cluster of items composing the project. A registration role for all the scans inside the group can be set: None,
Same as parent, Children move together. These roles are used to carry out a registration through groups.
The workflow is the following:
1. Find the groups of scans already registered.
There are two ways to create groups:

a. Automatically: When you automatically register scans with the Process environment, groups containing
just registered scans are created by default
b. Manually: You can manually create a group of already registered scans with that specific command.
In the example below, we identified two groups: A and B. We want to move all the scans in group B onto the scans
of group A, without losing the previously completed alignment of the scans.
2. Apply the Children move together registration role to the groups that you want to align. If a single scan is moved in
the 3D space, all the other scans move together, as a rigid system.
3. For each group, find a scan with a good overlap with one or more scans in other groups (in our example, these are
Scan2[A] and Scan3[B]).
4. Carry out a manual pre-registration among models to align Scan2(A) and Scan3(B).

Because the Children move together registration role of the groups permits a rigid and common movement of all the
scans of the groups, this procedure is enough to align all the scans.

Point cloud Georeferencing

This function allows you to georeference multiple scans. The workflow is divided in two steps:
• Reference points selection
• Scans geo-referencing
Reference points selection dialog
This dialog allows the selection of multiple reference points from the point clouds' target and center points. Target
points can be automatically or manually set using the targets registration dialog or targets editor dialog tools.
The buttons in the bottom part of the dialog have the following functions:
• Automatically remove duplicate points: If two or more selected points are within a 0.2 meters of each other,
only the higher-confidence point is selected. This option can be used to pick only one target point if the same
target was detected in more than one point cloud;
• Import selected: Uses only checked points for the geo-referencing step
• Import centers: Uses only the point clouds center points for the geo-referencing step
• Import targets: Uses only the point clouds target points for the geo-referencing step
• Import points: Uses all the listed points for the geo-referencing step.

Scans georeferencing dialog
This dialog allows the user to load external reference points for geo-referencing scans using previous selected target
points. The buttons in the bottom part of the dialog have the following functions:
• Show points selection: Opens the reference points selection dialog.
• Load external points: Imports external point from a textual point list.
• Load georeferenced centers: Imports the georeferenced center point of the point clouds, if present.
Georeferenced coordinates for the center point can be set automatically (depending on the format) or
manually when importing the given point cloud.
• Compute registration: Automatically finds the best matching points from the right and left lists and computes
the alignment between them. A registration report is shown at the end of the process.

Manage control points

This function opens the Target Editor.

Pose Transform
This function opens the Pose transform window.

Manage circular targets

This function opens the Target Editor.

Determine the cloud's vertical direction

This command is available in the Pose & Registration tools. It's used to define a plane that should be horizontal (e.g. a
floor) by double-clicking on it. This sets the vertical direction of the point cloud (z axis) as the normal direction of the
plane by rotating the point cloud.

Manual Positioning
This function opens the Manual positioning dialog.

Select in the Process Environment

You can find the Select command in the Tools menu or in the upper right side of the Top tool bar in the
Process environment.
You can switch between navigation and selection mode by pressing the spacebar.
The pick mode is the way to interact with the 3D data displayed in the rendering window.
The Select command helps the user to sample points in the currently visible frustum from the loaded point clouds. Both
3D and color are sampled. The shape of the region can be chosen with the selection mode menu.
Rectangle
Creates a rectangular selection in the current view of the point clouds. Left-click and drag the pointer in the
3D view to create the rectangle.
Polygon
Makes a polygonal selection in the current view of the point clouds. Left-click in the 3D view to create each
vertex of the polygon, then double-click to close the polygon.
Lasso
Makes a free-hand selection in the point cloud view. Left-click in the 3D view to start the polyline, drag the
mouse to draw the polyline, then release the mouse to close the polyline.
Complex selections can be performed by holding down special keys:

• Union: Shift
• Difference: Alt
• Intersection: Shift+Alt
There are two ways to delete points that are selected:
Delete inside
The points that fall inside the selected region are deleted from the view. They can be undeleted at a later
time.
Delete outside
The points that fall outside the selected region are deleted. They can be undeleted at a later time.
See also Restore Deleted Points to learn how to undelete points.

Measures & Notes in the Process Environment

In the Process environment you can take:
Distance This button allows you to measure a distance between any two points in the 3D scene.
After activating this button, left-click on the first point, drag the mouse to the second
point, and release the mouse. A result dialog will pop up displaying the distance between
the two points. You will be offered the option to save the distance as an annotation in
your project.
Save This function allows you to capture a snapshot of the 3D scene with its rendered models.
You can save the captured snapshot in a variety of image formats, including BMP, JPG, and
snapshot PNG.

Process Environment Settings Dialog
In this dialog, the user can specify advanced settings for the scans processing workflow.
See also Point Filtering & Clustering and Pose & Registration.

Wizard
This function opens the Scan Processing Wizard. Simply follow the instruction in the wizard to import, pre-process and
register point clouds. Clicking on More Settings causes the Processing Setting Dialog to appear.

Processing Settings Dialog

The Processing Settings Dialog is available by clicking More Settings in the Scan Processing Wizard.
It's split into several tabs:

• Importing
• Pre-processing
• Pre-registration
• ICP registration
• Bundle adjustment
These processes are managed by properly setting the parameters for each analysis step.
See details in Import, Point Filtering & Clustering and Pose & Registration.

Parse
The Parse window enables you to extract raw data collected by the Optech Polaris and import it into ATLAScan.
• Polaris Surveys: This pane shows the various surveys that have been loaded into this project. Selecting one of the
surveys puts its information into the Survey Info pane and lists its regions of interest in the Survey ROIs pane.
• Settings: Opens the Polaris Import Options window, where you can define which points to import, and whether to
colorize the points.
• Create Preset: Creates a preset base on the settings.
• Parse Selected: Begins parsing the selected surveys.
• Survey Info: Lists various information about the survey selected in the Polaris Surveys pane.
• Survey ROIs: A list of regions of interest (ROIs) in the selected survey.
• ROI Properties: Information about the ROI selected in the Survey ROIs pane.

Pose & Registration

ATLAScan includes various registration techniques:
• Define Project Reference System

• Manual Positioning
• Create/Edit UCS
• Point List Registration
• Pose Transform

Define project reference system
This dialog allows you to quickly define a new UCS by just specifying one point's coordinates in the old and in the new
coordinate system. You can input the coordinates in the “old” system in the left part of the dialog, and the “new”
coordinates in the right part. The button Define reference system will create a new UCS for you as desired.
You can simply copy the coordinates of a point (as below written) and paste them with Paste from clipboard button,
filling the empty spaces.
P25 1511.342 948.368 117.280
For more sophisticated methods of georeferencing, see also the Point Cloud Georeferencing tool.

Manual positioning (Adjust Pose)

It is possible to manually change the pose of an object by using the Manual Positioning (Adjust Pose) window. To open
the window, click Tools > Post & Registration > Manual positioning or press Ctrl+A. This window is enabled only if an
item in the Project Window is selected, in which case the bounding box of the current item is highlighted in bold yellow.
Insert the desired quantity of translation (meters) and rotation (degrees) and press the positive and negative buttons to
apply transformation.
If Object coordinates is selected, the rotation is around the origin of the object's coordinate system and the translation
is along the axes of the same system.
If UCS coordinates is enabled, the rotation is around the origin of the current UCS, and the translation is along the axes
of the same UCS.
See also the Pose dialog.

Create/edit UCS
This window is a toolbox for creating/editing a sphere. To create a new sphere, click the Create New button at the top
left. To edit an existing UCS, drag the UCS from the Project Window into the Drop a UCS here box, or click the Browse
button ( ) and select it from the list. Note that you cannot edit a UCS if it is being used as the current coordinate
system.
The following settings are available for creating or editing the UCS. In all cases, you can click Set as current UCS to
immediately start using the UCS.
Editing method
There are several ways to define the UCS. Selecting one of the options causes the window to display tools for creating
the UCS using that method.

From points
This creates the UCS based on selected points.
From current view

In this method, clicking Adapt UCS to Current View sets the current position of the 3D viewer as the origin of the UCS. It
also rotates the UCS so that X is to the right, Y is forward, and Z is upward in the current view.
Fit UCS to faces

In this method, you can define a UCS based on planes that you select from the 3D viewer. Click Define UCS horizontal
face or Define UCS XZ (side) face, then left-click in the point cloud, drag the mouse around the plane you want to use,
and let go. ATLAScan adjusts the current UCS to fit the selected plane, depending on whether it is horizontal or vertical.
From an item's pose

In this method, you can use an object in the 3D viewer (such as a cone or a camera) to provide the UCS so that its
orientation and origin fit the center of the object. To select an object, double-click on it in the 3D viewer or drag it from
the Project Window into the Recipe Window.
From points registration

In this method, clicking Open Points Registration opens the Georeferencing from point list window.

Georeferencing from point list
This dialog allows you to register a list of points, taken from a text file or from the point list window, against another list
of points that you consider as reference. In particular, these last points can be geo-referenced.
The left half of the dialog regards the moving points. You can load them from a text file with the button Load moving
points; in this case you are invited to select a text file and then the parse point list dialog appears. Alternatively, you can
copy the points listed in the point list window via the button Copy moving points from point list window. Clicking OK
makes the moving points appear in the left table.
The right half of the dialog works handles reference points. Clicking Load reference points on the top right loads the
points from a text file, which then appear in the right table. To start the registration, the left table has to contain as
many points as the right table.
When you have loaded the points, press Compute on the bottom right. A dialog appears, asking whether you want to
register the points by coupling them according to their labels, or by trying out all the possible pair combinations to find
the best. The first option is much faster but it assumes that you are sure about how to match your points. After you have
selected either Match names or Best fit, you can refine and make use of the results of the registration in the Registration
report dialog.
This tool registers a list of points against another list of points. For a general overview of ATLAScan's registration tools,
see Pose & Registration.

Registration report dialog
This dialog allows you to refine and exploit the result of a registration. This dialog is used as final step of the following
procedures: single-scan georeferencing, georeferencing from point list, and platform calibration.
• On top of the dialog, the mean registration error is shown, along with the editable error threshold.
• In the central table of the dialog, the point pairs are shown along with their associated registration errors. Errors
above the error threshold are displayed in red. You could consider the corresponding pair as an outlier. If you
uncheck an outlier pair by pressing the checkbox on the left, then press Update registration on top right, the
registration will be recomputed without the outlier and the mean registration error will improve.
• In the bottom part of the dialog, there are options on how to make use of the obtained registration. The
option Copy transform in clipboard is useful for having a backup of the registration transform to use on project
entities that will be created at a later time. The option Make backup of poses of affected project items before
applying the transform will save the poses of the affected project entities in the Exports folder of the project, these
poses can be restored later via the matrix transform tool to undo the effects of the current registration. To access
this tool, go to the project window, select the entities you want to register, right-click, and choose Registration >
Matrix transform from the context menu.
• The button Apply transform to all project on the bottom extends the effect of the registration to all the project
entities. All your project items (including annotations) will be moved according to the roto-translation defined by
the registration. The button Apply transform to a project selection will pop up a dialog to select a subset of the
project items. Only those selected items will be moved.

Pose Transform
The Pose transform command allows you to apply to your project entities the transforms coming from registration
functions.
In pre-registration, registration, and geo-referentiation, you can copy in the clipboard the transform that results from
the computation.
• The Apply registration transform button searches the clipboard for such a transform. If one is found, the transform
is applied to multiple models.
• The Apply inverse transform button undoes the effect of the first button. It reads from the clipboard a registration
transform and applies its inverse to the selected project entities. this is useful if you don't like the results of a
registration.
• The Define transform matrix button allows you to manually define a transform matrix, which is post-multiplied to
your project entities. The transform must be defined as referring to the current UCS. A common situation in which
this function is needed is when you have done a point list registration and copied the resulting transform to the
system clipboard. In this case, to apply the transform press Define transform matrix, then Paste
matrix and Commit. The selected project entities will be moved according to the registration transform matrix in the
clipboard.
• The Apply pose from file button allows you to import the POSE files you export with the "Export pose" function,
which is reachable from the context menu of any project item. It is designed to save the positions of your project
items for backup. Therefore, you can use the Apply pose from file function to move your models to positions you
saved in backup files.

Note: The pose matrix exported by the "Export pose" function is different from the pose matrix displayed in the Pose
Dialog. The "Export pose" function is designed to provide to let you make a backup of the items' positions and to restore
them at a later time in case something goes wrong. Now, if "Export pose" were to save the pose in reference to the
current UCS (that you see in the Pose Dialog), the pose would be lost if you accidentally moves or deletes the UCS.
Therefore, "Export pose" exports a pose in reference to an internal, hidden, never-changing ATLAScan global reference
system.
See also Pose Dialog.

Pose Dialog
In ATLAScan, the 3D position of each object (except annotations) is described by the POSE matrix (transformation matrix)
that you can open pressing the pose button in the Property Browser Window or by using an item's context menu (Pose
& Registration > Pose).
You can manage different UCS systems (local or global/geo-referenced). A transformation matrix defines how a project
entity is located and oriented with respect to the current UCS. Therefore, if you have knowledge of transformation
matrices, it is possible to perform specific and precise translations and rotations on a project entity.
A transformation matrix is a 4x4 matrix that specifies how an object O is located and oriented with respect to a global
coordinate system. The first, second and third column of the transformation matrix contain respectively the versors of
the X, Y, and Z axes of the system anchored with the object O. The fourth column contains the translation vector of O's
origin with respect to the UCS' origin. The fourth row of the translation matrix is always (0 0 0 1).
See also Manual positioning (Adjust Pose).
Restore a Pose
You can restore an object position in case of registration and geo-referencing error once the Pose file is saved. To save
the Pose you have to export it by right-clicking on it in the Project window and selecting Export > Export Pose from the
context menu.

The software automatically creates a folder that contains the pose exported. This folder is saved in the Exports folder
and has the name PoseFiles plus the date and time of its creation. Apply the pose file by selecting the desired objects,
right-clicking on them, and selecting Pose & Registration > Pose transform, which opens the Pose transform window.
Select Apply pose from file and then select the folder containing the pose you want to restore.
Advanced Options
This dialog is designed for advanced users, since it requires specific knowledge of 3D computer graphics, in particular
knowledge about transformation matrices.
In the dialog there is a central 4x4 table displaying the transformation matrix of the current project entity. On the sides,
the buttons Post-multiply and Pre-multiply are located.
If M is the current matrix, a pre-multiplication by matrix P is defined as

M’ = M ∙ P
If M is the current matrix, a post-multiplication by matrix P is defined as

M’ = P ∙ M
Any number of pre-multiplications or post-multiplications to the current matrix are possible.
Confirm any modification of the matrix by pressing Commit on bottom left of the dialog. Press Revert to discard any
modification. The upper-right Close button (x) has the same effect as Revert and hides the dialog.

To speed up the definition of the matrix the following operations are available:
• Copy in clipboard: Copies the current matrix to the clipboard, which is useful for exporting it or pasting it to
another model.
• Paste from clipboard: Pastes the matrix from the clipboard. Confirm the change by pressing Commit.
• Euler form: Define the transformation as rotation plus translation. The rotation is defined as a sequence of
rotations along the local coordinate frame axis. The pitch angle is around X, heading is around Y, and roll is
around Z; these are measured in degrees. If the sequence is YXZ, then first the rotation around Z(Roll) is
applied, then X(Pitch), then Y(Heading). Press OK to have ATLAScan compute the homogenous transformation
matrix.
• Axis-angle form: Defines the transformation as rotation plus translation. The rotation is around the desired
axis (X,Y,Z) direction by amount Angle, with the right-hand-curl rule. Press OK to have ATLAScan compute the
homogenous transformation matrix.
• Zero: Clears the matrix to zero
• Identity: Defines the matrix to be the identity
• Invert: Inverts the current matrix
• Transpose: Transposes the current matrix
For example
Rotation around X axis by 90° Rotation around Y axis by 90° Rotation around Z axis by 90°
1 0 0 0 0 0 1 0 0 -1 0 0
0 0 -1 0 0 1 0 0 1 0 0 0
0 1 0 0 -1 0 0 0 0 0 1 0
0 0 0 1 0 0 0 1 0 0 0 1
Moving objects with the roto-translation matrix

Copy Translation button: Copies the object translation (column 3). Use this option to translate an object on these
coordinates.
1. Click Copy Translation.
2. Open the pose of the object to move.
3. Click Paste translation.
To translate an object according to specific coordinates, type the desired coordinates in the column 3 (first line stands
for X, second Y, third Z). The translation will be made according to the object’s UCS reference system.

Selection & Fitting Tools

In this section you can select portions of point clouds and fit them with certain features.
• Select
• Point selection with polyline
• Sample to new cloud
• Fitting

Select
In this section you can learn how to select portions of one or more point clouds.
You can find the Select command in Tools > Selection and fitting tools menu, or in the upper right side
of the top toolbar.
You can switch between navigation and selection mode by pressing the spacebar.
The pick mode is the way to interact with the 3D data displayed in the rendering window.
The Select command helps the user to sample points in the currently visible frustum from the loaded point clouds. Both
3D and color are sampled. The shape of the region can be chosen with the selection mode menu.
Rectangle
Creates a rectangular selection in the current view of the point clouds. Left-click and drag the pointer in the
3D view to create the rectangle.
Polygon
Makes a polygonal selection in the current view of the point clouds. Left-click in the 3D view to create each
vertex of the polygon, then double-click to close the polygon.
Lasso
Makes a free-hand selection in the point cloud view. Left-click in the 3D view to start the polyline, drag the
mouse to draw the polyline, then release the mouse to close the polyline.
Complex selections can be performed by holding down special keys:

• Union: Shift
• Difference: Alt
• Intersection: Shift+Alt
During points selection you can choose how select points:

Deep Selection
Samples the selected region. All points that fall into the current video selection, even if they are hidden by
more prominent objects, are selected.
View Selection
Sample the selected region from the depth buffer. This means that only the visible points will be selected.
The accuracy of the points depends on the current frustum depth range. The number of points is exactly the
one of the current screen resolution.
Suggestion: Get close to the object you want to select and keep the maximum order of magnitude between
the near and far planes in view cut planes under 4.

[See View Parameters Dialog]
Reset samples
Discards all the current samples.
There are two ways to delete the selected points:
Delete inside
The points that fall inside the selected region are deleted from the view. They can be undeleted at a later
time.
Delete outside
The points that fall outside the selected region are deleted. They can be undeleted at a later time.
See Restore Deleted Points to learn how to undelete points.
Finally, there are also several tools for hiding points:
Hide inside
The points that fall inside the selected region are temporarily hidden.
Hide outside
The points that fall outside the selected region are temporarily hidden.
Restore hidden points

Shows all points again.
Toggle visibility
Makes all hidden points visible, and makes all visible points hidden.

Point selection with polyline

This is a tool for selecting points using a 3D shape, in particular a polyline.
Drag the point cloud to be used as source of your selection into the recipe window. Likewise, drag the polyline you want
to use as the contour to select points from your cloud. Finally, specify a planar reference. This can be the current view or
a custom plane.
The procedure works as follows: the polyline is projected on the planar reference you chose. If the polyline is not closed,
it is automatically closed by connecting the last point to the first. Then, the polyline projected on the plane delimits a 2D
closed region. The points of your cloud are also projected on the plane: the ones falling inside the projected polyline are
marked as selected, the ones falling outside are marked as not selected.
If no points were selected, ATLAScan gives a warning message and stops the procedure. If any points were selected,
ATLAScan creates a new unstructured point cloud called Selection of <input-point-cloud-name> and stores it as a new
item in the project.

Samples to new cloud

You can find the command Samples to a new cloud in Tools > Selection and fitting tools.
It can be difficult to manage large projects comprised of a large number of scans. When it’s possible to work on a small
region it’s more advisable to extract a point cloud’s portion to lighten the processes.
Starting from a selection of points (see Select), the command Samples to a new cloud creates a new point
cloud called Sample and stores it as a new item in the project.
If you selected the points with Deep Selection command, a new Unstructured point cloud is created.
If you selected the points with View Selection command, a new Grid (structured) point cloud is created.

Selection and fitting

After selecting a sample of points with the Selection Tools, you can fit a plane, cylinder, sphere, circle or cone around
them using the tools in this menu. various shapes around them using the tools in this menu.
You can also fit a shape over a list of points by clicking Fit shape to point list. If you do not have any points currently
selected in the Project Window, ATLAScan will ask you to select some points, then it opens the following window.
In this window, you can adjust the selected group of points, adding new points by clicking and removing removing
selected points by clicking . Adjust the fitting settings by selecting the type of shape you want to create in the Fitting
settings dropdown menu. Finally, click on one of the New Shape buttons in the top bar to fit the shape to the points.

Points filtering and clustering

This category of tools includes functions dedicated to work with point clouds , to enable other processes and further
results. Most of these tools work on any point cloud, some of them work only on grid point clouds.
• Pre-process Grids
• Restore raw data
• Restore deleted points
• Edit 2D (Grid point clouds only)
• Point clustering
• Make single cloud
• Fill holes (Grid point clouds only)
• Hide black points
• Remove duplicate points
• Resample
• Simplify points (Grid point clouds only)
• Extract edges (Grid point clouds only)

Pre-processing Grids
During preprocessing, ATLAScan applies a set of algorithms to the range scans, which extract information that is needed
during further processing of the data.
• Load adds an external grid point cloud

• Scan subfolders automatically finds all grids in a folder and its subfolders.
• Remove removes undesired grids. It is possible to process grids present in the current project by selecting Pre-
process from the grid item's context menu in the Project Window. In this case, when the processing is finished, the
grid is unloaded from the project to free memory. You must force a reload to refresh the rendering.
• Start begins processing. This creates a separate thread for each grid, so if multiple CPUs are available the processing
is sped up proportionally.
• Select all applies all steps consecutively to the selected grid. A step algorithm can be changed by clicking the
algorithm name, a combo box of available algorithm is shown. A step can be skipped by unchecking it.
• Advanced modifies the pre-processing parameters. You must select one or more rows in the list of models with Ctrl
or Shift, then select a processing step. The parameters are updated for all the selected models.
If Use backup is checked, the pre-processing reads the backup data. Note that ATLAScan never overwrites the original
data. The first time the pre-processing begins, a backup of the data is created. By default the pre-processing pipeline
starts from the data in its current state, but selecting use backup causes it to use the backup data.

Noise Removal
The noise removal steps have the following options.
Range & Reflectance Gate

• Range min/max [m]: lower/upper threshold of the range. All pixels outside this range will be filtered.
• Reflectance min/max [0-1]: all measurements that have an intensity value outside the given interval will be filtered.
Median Filter
• Mask border [pix]: The kernel of the filter is a square mask (of side 2∙border+1) centered at the current pixel.
Mixed Point Filter

This is useful as a relative depth discontinuity filter.
• Min incident angle [deg]: If the line of sight from the origin of the grid to the point has incident angle to the local
surface less than this value, the point is filtered.
Compute Normals
The compute normals tasks have the following options.
Local Plane Fitting

This task computes the local surface tangent plane for each point, based on the neighborhood of the pixel.
• Neighbourhood size [pix]
• Compute also curvature: Set TRUE to compute it.
• Adjust normals' verse in handheld scans: Set TRUE if processing handheld scans.
Edge Detection
The edge detection tasks have the following options.
Depth & Orientation Discontinuity

This task computes geometrically significant line features from the point cloud data. The extracted edges are stored as a
bit flag for each point. Two types of edges are extracted:
• Depth discontinuities (or jump edges) that occur where the scanner hits an occlusion and therefore the measured
range jumps from a foreground to a background value.
• Orientation discontinuities (or crease edges) that occur where the object has a sudden change of its surface
orientation.
This tasks has these options:

• Min depth discontinuity to flag [m]: Absolute depth discontinuity to filter.
• Min orientation discontinuity to flag [deg]: The current point is marked if its normal differs from the adjacent ones
of at least this angle.

Compute Confidence
The computer confidence tasks have the following options.
Weighted Average
This task computes a confidence value for each measurement, which is a measure for the reliability of the given range
measurement. The accuracy does not only depend on the type of scanner used, but also on the following factors
(amongst others):
• The incident angle between the laser beam and the tangent plane of the target
• The distance to the target
• The material of the object and therefore the intensity of the reflected signal
The confidence value is computed as a weighted sum of the surface normal, the range value and the reflectance value.
This task has these options:

• Scale factor: The confidence value calculated for each pixel is multiplied by this scale, enabling you to decrease or
increase the weight for a given scan manually, for example because a scanner with higher accuracy has been used
during acquisition.
• Range min/max [m]: The range is weighted by normalizing it within this interval.
• Weight of range/reflectance/inclination: Modify these weights as desired.

Restore Raw Data

Input data: At least one unstructured point cloud must be selected.
This functions takes as input a set of point clouds and undoes any preprocessing, deletion and editing operations that
may have been performed on them. Its main use is to guarantee that you never lose data: if you are not happy with an
editing operation or with the result of a filter, you can always go back to the clouds' raw data, as if they were just
imported.
Therefore, when applying this function:

• All deleted points are visible again
• Points “confidence” layer is deleted
• Points' normals are deleted
• Points invalidated by the range and reflectance gate are valid again
• Points invalidated by the outlier remover filter are valid again
• Points invalidated by the mixed point filter are valid again
• The effect of any application of the median filter is undone
• Any flags indicating that points are depth or orientation discontinuities are deleted

Restore deleted points

Sometimes point may be deleted by mistake. In this case, ATLAScan has a way to recover deleted points, even at
different times, valid for both Grid and Unstructured point clouds.
To undelete all previously deleted points, select Tools > Points Filtering & Clustering > Restore Deleted Points from the
menu, or right-click on a point cloud in the Project Window and select Points Filtering & Clustering > Restore deleted
points.
If you want to undelete only the points in a defined area, use the command Predefined Views > 2D view from the menu
(also available under Filtering & Clustering > 2D view in the context menu of the grid). This only works for grid point
clouds. Select a region containing the portion previously occupied by the deleted points using a pick mode and then click
Undelete.
See also Edit 2D Grid.

Edit 2D Grid Window
This window shows a grid point cloud in its 2D representation. Like in a geographic 2D map of the world, the grid point
cloud is represented as an image where, for each point, the x coordinate is the yaw angle (longitude) and the y
coordinate is the pitch angle (latitude) that the laser had while surveying that point. In this window you can select,
delete and undelete points with several functions.
To interact with the image

• Zoom in/out: Alt + mouse wheel
• Fit the image in the window: Alt + Middle Mouse Button (MMB)
• Reset to original size: Alt + Right Mouse Button (RMB)
• Translate the image: Shift + LMB drag
• Rotate image 90 degrees: Space bar
• Show/hide zoom window: Z key
• Move zoom region inside global view: Ctrl + LMB
• Perform union of regions: Hold Shift and select area
• Perform subtraction of regions: Hold Alt and select area
• Perform intersection of regions: Hold Shift+Alt and select area

Pick mode
• Point: Pick a point. If the AutoCAD link is enabled, the 3D coordinates are sent to AutoCAD
• Rectangle: Left-click on the first point, dragging it to the opposite corner, then release the mouse.
• Polygon: Left-click on each point of the polygon. Close the polygon by double-clicking.
• Lasso: Left-click to start the polyline and drag the mouse to draw, then release the mouse to close the
polyline.
Edit operation
• Copy: Creates a new grid point cloud from the selected region.
• Cut: Deletes the selected region.
• Undelete: Undeletes the points inside the selected region.
• Mirror: Inverts the grid along the width. Only the ordering of the grid is changed, the 3D points are
untouched.
• Flip: Inverts the grid along the height. Only the ordering of the grid is changed, the 3D points are untouched.
• Transpose: Transposes the grid. Only the ordering of the grid is changed, the 3D points are untouched.
Show
• Color type: Displays the grid with the selected color type. Press the right arrow to show a menu of available
commands for the current color type.
• Histogram: Shows the histogram tool, which allows you to optimize the contrast by histogram stretching
(available only if the model is loaded in memory and if the current color type is 1f or 1d, i.e. a high-dynamic
single-channel point color).
• Colors mapping: Remaps point colors to a pseudo-colored scale to improve the dynamic range. This is useful
if the histogram stretching is not sufficient, and can also be used for range segmentation and inspection. It is
available only if the model is loaded in memory and if the current color type is 1f or 1d (i.e. a high-dynamic
single-channel point color).
• Global coordinates: Displays the 3D global coordinates of the point (i.e. using the pose of the model).
Otherwise, the local model coordinates are used.
• Save image: Saves the current image to file.

Point clustering
This dialog allows to lump together an arbitrary set of point clouds, structured or not, in an unstructured point cloud.
The resulting cloud, however, will not contain all points from the input clouds, but only those needed to guarantee a
fixed 3D density of the points.
Regularize the density of your clouds

Laser scanner data can have quite irregular 3D densities. For example, objects very close to the laser scanner may be
sampled at one point each half millimeter, but objects far away from the scanner may be sampled at one point every
30/40 cm. This tool allows you to regularize the 3D density of your clouds. The resulting regularized cloud can then be

used for meshing or for exporting in CAD tools for further analyses.
To activate this dialog, you can select a set of point clouds from the project window, right-click on them, and select Point
Filtering & Clustering > Point clustering from the context menu, or click Tools > Point Filtering & Clustering > Point
clustering in the menu bar.
The selected clouds appear in the top list of the dialog. Below, the common color layers across the clouds are found. The
cloud that results from the clustering will contain only the color layers that are present in all input clouds and that are
chosen. Point normals will be transformed so that they will be correct in the resulting cloud.
Options
In the lower half of the dialog, some options are present. The first is the desired 3D density, expressed in terms of the
minimum point-to-point distance. You can also select the strategy for choosing the representative point of each 3D cell.
This procedure will efficiently exploit all CPU cores of your PC.

Make single cloud

(Single Point Cloud Clustering dialog)
This dialog allows you to lump together in an unstructured point cloud an arbitrary set of point clouds, structured or not.
If you want to cluster clouds while excluding duplicated or unneeded points, consider using the Point clustering tool
instead.
To activate this dialog, you select a set of point clouds from the project window and select Point Filtering & Clustering >
Make single cloud from the context menu or from the tools menu.
The selected clouds appear in the top list of the dialog. Below, the common color layers across the clouds are found. The
cloud that results from the clustering will contain only the color layers that are chosen from those present in all input
clouds. Point normals will be transformed so that they will be correct in the resulting cloud.
Below in the dialog, you can specify a sub-sampling step if you do not wish to include all points of the input clouds in a
new cloud.
The clustering procedure will efficiently exploit all CPU cores of your PC.

Fill holes
Input data: Exactly one grid point cloud must be selected.
This function takes as input exactly one grid point cloud, and tries to replace any invalid point in the cloud with a value
averaged from the point's neighbourhood in the cloud's structure. After activating this tool, ATLAScan asks you for the
desired mask border M. The search neighbourhood of each point is defined by the (2M + 1) * (2M + 1) points located
around the given point in the cloud's structure.
This tool is useful for covering small points in a cloud's structure to get more connected meshes using the
multiresolution mesh.

Hide black points

This functions takes the current color from each of the currently selected point clouds and invalidates all the points in
the cloud that are colored in full black (RGB components [0 0 0]).

Remove duplicate points

This function invalidates any point in each cloud that has exactly the same coordinates of another point in the cloud.
After its conclusion, this function prints to the Log window the number of removed points per cloud. Once removed, the
duplicated points cannot be restored in any way.
This function is automatically applied to all point clouds while importing them, so it is not normally necessary to run it
manually.

Resampling dialog
This tool allows you to resample point clouds. It works on any set of grid point clouds or unstructured point clouds.
On top of the dialog you are asked to input the subsampling factor s. A new point cloud will be created taking one point
per s from the original cloud. If you are resampling grid point clouds, the resampling operation will take into account the
structure of the grid. Therefore, the grid structure will be divided in cells with size square_root(s) by square_root(s), and
one point will be taken for each cell. To clarify this for grid point clouds, the dialog shows you a message with the original
and resulting grid size, as in the figure above.
Optionally you can hyperlink the subsampled scan with the original high-resolution scan. This allows you to work with a
lot of data by doing operation on the resampled version of the data (which is often much smaller than the original).

Simplify points
Input data: At least one grid point cloud must be selected.
Output: A new unstructured point cloud is created for each structured cloud in input.
For each point cloud, the algorithm determines the most relevant points with regard to shape description, and saves
them into a new unstructured point cloud. These resulting clouds work as compact representations of the original
structured ones. They can be hyperlinked to the original clouds, so that you can open the original data on demand.
The function works by creating a mesh in the background of each input cloud, and by inserting the mesh's vertices into
the output cloud.

Extract edges dialog

This tool allows you to extract the edges of a grid point cloud as polylines. You can activate this function from the top
menu by clicking Tools > Point Filtering & Clustering > Extract edges, or through the contextual menu of any grid point
cloud.
You can select which discontinuities in the 3D point cloud you want to mark as edges. These can be the Depth
discontinuities (points that are neighboring in the grid structure but far away in 3D position), the Orientation
discontinuities (points that are neighboring in the grid structure, close to each other in 3D position, but having very
different normals), or both. You can also mark as edges all points previously marked with the User mask. This is a mask
reserved for advanced operations, such as the mask filter.
The remaining parameters define the way edges are extracted:

• Max search distance: The distances (in pixels) in the grid's structure (range image). It determines how far away in
the grid structure two points belonging to the edges must be connected in the same edge.
• Min edge length: All the edges composed of fewer points than this parameter will be considered as noise and
discarded.

Meshing
In this section you can first learn how to create triangle meshes.
ATLAScan provides several typologies of mesh techniques. Different numbers and types of point clouds can be used as a
basis to create a mesh, depending on the meshing technique. You can access these commands from Tools > Meshing,
from Mesh tools in the top toolbar, from Tools > Meshing in the menu bar, or from the context menu of point
clouds.
On On
Grid Unstructured Suggested for…
Point Clouds Point Clouds
Multiresolution Mesh Useful for getting a
Fast meshing technique that creates light meshes well-defined mesh
that may have holes in some situations.  quickly from a single
structured point cloud,
(one or
with a good
more)
quality/computational
time ratio
Mesh from predefined view Useful for façades
Relatively slow meshing technique that creates  (using orthocamera)
convex meshes without holes. It’s a view-dependent,
high defined mesh (each point is a vertex).
 (one –
single or
and tunneling (using
cylindrical camera)
(one)
clustered
point cloud)
3D Mesh Useful for convex
Approximate 3D meshing technique that is not view-  surfaces
dependent and takes as constraints the points'
positions and orientations (normals).
 (one –
single or
(one)
clustered
point cloud)
Topographic Mesh Useful for land survey
The algorithm designed for DTM models. It quickly   and mining
produced a light, smoothed, watertight mesh that is (one or (one or
useful for isoline and volume calculation. more) more)
See the section Mesh Editing to learn all the editing meshes techniques.

Multiresolution Meshing
Multiresolution Meshing is possible only with Grid point clouds. If you want to generate a mesh from an Unstructured
point cloud, first transform it in a Grid point cloud by:
1. Using the Virtual scan tool
2. Clicking Tools > Selection and fitting tools > Samples to a new cloud after selecting the unstructured points
with Selection tools and the View Selection command
NOTE
Before meshing you need to pre-process the grid. During the meshing the software will use the edges calculated at the
pre-processing stage:
• Angular discontinuity: Set the angle [deg]; the

higher the angle value, the higher the detail of the
calculated edges. Close to the angular discontinuity
you will have more detailed triangles
• Depth discontinuity: Set the minimum distance [m]
two points must have to be considered as belonging
to two separate objects  no triangles are created
between two far points. The lower the distance
value the more holes are filled.

The Multiresolution Mesh tool is accessible from Tools > Meshing > Multiresolution Mesh, from the
corresponding button in the top toolbar, or from the context menu of grid items in the project window.
• Press Load to add an external grid point cloud or Scan subfolders to find automatically all grids in a folder and its
subfolders. Press Remove to remove undesired grids. Note: It is possible to process grids present in the current
project by selecting Meshing from the context menu of the grid item in the Project Window. In this case, when the
processing is finished, the grid is unloaded from the project to free memory. You must force it to reload to refresh
the rendering.
• Press Start to run the processing. This task creates a separate thread for each grid so if multiple CPUs are available
the processing is sped up proportionally. The meshing process is applied consecutively to the selected grid by
pressing the Select All button.
• To modify the meshing parameters, press Advanced and select one or more rows in the list of models with Ctrl or
Shift. The parameters are updated for all the selected models.
• It is possible to save the meshing parameters by clicking Save and load them later for other scans or projects using
the Load button. You can also or click Reset to default to use the default values.
Triangulation
The triangulation algorithm can be changed by clicking (three times) the algorithm name, so a drop-down box of
available algorithms is shown. These algorithms are described below

Uniform grid triangulation

This algorithm connects all the 3D valid points with triangle meshes to build the range surface, without simplification.
The algorithm checks if 2 points are divided by a depth discontinuity, in that case no “false” triangles are created. It has
the following parameters:
• Save as: Enter the file name of the resulting mesh.
• Color mapped: Select the color assigned to the vertexes of the mesh (e.g. Reflectance, Color, Inclination,
Confidence, …)
Multi-resolution grid triangulation

This algorithm creates simplified (lighter) triangle meshes in “flat” areas, and dense triangle meshes in geometrically
complex areas (such as close to edges or curvatures), satisfying an approximation error threshold. This algorithm has the
following parameters:"
• Save as: Enter the file name of the resulting mesh.
• Max triangle size [pix power of 2]: 2 [number of points],, it’s the size of initial triangle; the value you can set is [number of
points] you want to skip in the flat area. The higher the value, the more the mesh gets simplified. As example, a
value of 4 means that a triangle will be created for a square 2^4=16 pixels wide. If the approximation error is too
high, the triangle will be subdivided in smaller pieces.
• Color mapped: Select the color layer you want to map.
• Use orientation discontinuity: The pre-processed orientation discontinuity is used to drive the subdivision process
in order to preserve features such as edges. Set to TRUE to have more triangles around the edges.
• Min triangle aspect ratio: The minimum accepted triangle aspect ratio. If the triangle is too thin, it will be
subdivided into smaller pieces. This test is performed only for the initial triangle. Note: The aspect ratio of a triangle is
the ratio of the inradius to its circumradius). The aspect ratio of a triangle lies between 0 and 0.5. For triangles with an angle near
to zero, the aspect ratio is 0; for equilateral triangles, the aspect ratio is 0.5.
• Accuracy [m]: The maximum approximation error of the mesh. The higher the value, the fewer triangles will be
produced.
• Create texture for color mapped: Generates a texture from the selected Color mapped on the mesh. Set to TRUE if
you do not want the color layer to lose resolution during the mesh simplification.
The images below show the results of two different set of parameters on the same original dataset.

See also other Meshing techniques.

Meshing from predefined view

This meshing technique provides watertight meshes. It assumes that the desired mesh is such that each point of it has
one and only one projection according to a predefined view. In other words, it assumes that a perspective, orthographic,
cylindrical or spherical projection exists such that the desired mesh never hides parts of itself if seen from that
projection.
In order to create the mesh you need:

1) An unstructured or structured point cloud: Drag the point cloud you want to mesh from the Project window.
2) A view: you need to indicate also which view should be used to build the mesh. The current view can be used,
or you can use any camera from your project (see below for some examples).
When the cloud and the viewpoint are selected, click Ok and wait for the end of the meshing. You can also cancel the
meshing during processing. At the end of meshing, a dialog shows you the new mesh's name and properties.

Creating a mesh from an orthographic view: a Creating a mesh from a predefined cylindrical view:
façade a tunnel
This algorithm works by projecting all the points of the cloud on a 2D surface: the near plane of the selected projection.
Then, these points in 2D are meshed with a standard 2D meshing algorithm: Delaunay. The resulting mesh is then
projected back into the 3D space.
This meshing technique is interpolative: each point of the input cloud is considered as vertex of the output mesh.
ATLAScan's Multiresolution and Topographic meshes are, instead, approximative meshing techniques: some points may
not be mesh vertices if they don't significantly modify the surface's shape. For this reason, this interpolative technique is
generally slower than approximative ones, and produces heavier meshes. However, it always produces watertight
meshes.
Heavy meshes or meshes with wrong shapes can be simplified and edited in the mesh editor.

3D Meshing (Poisson reconstruction)

After selecting point samples with Selection Tools, a surfaces fitting is possible.
This tool allows you to reconstruct a triangle mesh from any point cloud. Unlike uniform or multiresolution meshing, this
technique does not need the input cloud to be structured. Also, unlike meshing from predefined view and topographic
meshing, this technique is not view-dependent. It reconstructs a connected, watertight surface taking as constraints the
points' positions and orientations (normals). It is suggested to reconstruct a convex surface. If you need to work with a
set of point clouds (structured or not), it is better to lump them together in an unstructured point cloud using the Level
3D density of clouds command.
In the 3D Meshing Dialog you can manage some parameters to obtain a best fitted mesh.
• Adding color: In the first parameter box of the dialog, you can select whether the output mesh should be colored or
not, and which of the input cloud's color layers should be used.
• Specifying how many details: While meshing from predefined view is interpolative, this reconstruction technique
is approximative: it creates a triangulation that approximates the surface described by the points. In the second
parameter box, you can specify how many triangles you want to use for your output mesh. The more the triangles,
the finer the details that will be reconstructed, and also the heavier the mesh file and rendering time.
• Taking into account cloud smoothness: In the third parameter box, you can take into account the cloud
smoothness. If your cloud represents clean construction data, then select Smooth. If, however, your cloud contains
noisy territorial data, then select Noisy.
• Closing holes and removing isolated components: In the last parameter box, you can specify the size of the holes
that are going to be closed in the mesh, and the size of the “islands” (small, disconnected components that may
have been created). The size is computed as ratio between the area of the hole/island and the area of the whole
mesh.

Topographic meshing
This function implements topographic meshing of a set of point clouds. This procedure takes a set of point clouds and
creates a DTM regularly sampling the clouds. The resulting mesh is watertight if the default parameters are used, and it
is colored according to the altitudes: points at minimum height are in red and points at maximum height are in violet,
passing through all hues.
The user can also input a polyline to determine the points to be meshed (seen from a nadiral view or from a user-
specified orthocamera). It is strongly recommended to input a polyline in order to help ATLAScan concentrate only on
the useful points and to speed up the process.
If the user doesn't specify any orthocamera, one is created so that it is nadiral (oriented towards -Z axis in the current
UCS) and it contains the polyline. Once the user inputs the clouds and (optionally) the polyline, clicking OK starts the
meshing.
The procedure is composed of the following stages:

• A virtual scan is performed from the (nadiral) orthocamera, to uniformly sample the point clouds.
• Spikes in the resulting samples are smoothed using a median filter based either on a neighborhood of given size, or
on a window of fixed size.
• Successively, a polynomial interpolation (ordinary gridding) is used to fill holes. This eliminates holes in the samples
and guarantees the result mesh to be watertight.

• A part of the obtained grid point cloud is selected using the polyline and the orthocamera. The selection with
polyline allows the user to specify the region that he/she wants to be meshed.
• A median is applied to the resulting grid, to smooth it out.
• The resulting mesh is obtained by multiresolution meshing, filling all holes.
• The mesh is saved in the project, under the name Topographic mesh of <n> clouds.
This procedure is relatively fast and is useful especially for terrain. It outputs a mesh that is watertight, light and smooth
(without spikes). The procedure can be customized by clicking the set parameters button.
See also other Meshing techniques and Topographic meshing parameters.

Topographic meshing parameters

In this dialog you can set the parameters of the topographic meshing procedure.
This dialog shows from top to bottom the four stages of the topographic meshing procedure. You can customize the
parameters of each stage, and even decide to skip stages 2 and 3 by unchecking the respective boxes.
• Stage 1: DTM generation. The terrain sampling step is the fundamental parameter of this procedure. It is important
to make sure that its value is appropriate for the dimensions of the desired model. For example, if you want a mesh
of a terrain of 20 m², you should not use a sampling step of 0.5 m, or you will get too few samples. A smaller step
will produce more accuracy and heavier meshes. A bigger step will give less accuracy and also less memory occupied
by the final mesh.
• Stage 2: despiking. Despiking is very useful for deleting noisy data (e.g. vegetation) and therefore allows more
precise calculations of volumes and isolines on the resulting mesh. Despiking is done with a median filter running on
the grid point cloud obtained in the former step. As the dialog shows, the median filter can run on the N nearest
neighbors of a sample, or on a window of samples of fixed size, regardless of how many samples are actually in the
window.
• Stage 3: hole filling. This is performed by taking, for each missing sample in the grid point cloud, the <neighborhood

size> samples closest. Polynomial interpolation is used to decide the height position of the missing sample. After the
hole filling, another median filter is run on a square of samples of fixed size. The user can save the grid point cloud
obtained till here by checking Save grid to project before meshing. This can be useful when (for example) you need
to mesh the grid using a color not related to the altitudes.
• Stage 4: meshing. At the end, the grid point cloud is meshed using one of the two techniques described here. Since
at this stage the grid point cloud should be smooth, regular and without holes, there is no reason to change these
parameters except for very particular cases.
You can use the Reset to defaults button on bottom left if you are not certain how these parameters will affect the
mesh. However, topographic meshing is so fast that it is worthwhile to try many combinations of parameters and
compare the results.

Mesh Editing
In this section you can first learn how to edit triangle meshes to obtain a better result from your data. ATLAScan offers
several functions for editing meshes:
Mesh Editor An editing environment that lets you perform advanced operations such
as hole-filling, border detection, triangle and vertex editing, surface
smoothing, decimation, and ridge and valley extraction.
Mesh selection from current Cuts a portion of a mesh using 2D video selection tools on the current
view point view.
Mesh selection with 3D Cuts a portion of a mesh using an input mesh, a 3D polyline and a
polyline viewpoint
Make single mesh… Lumps an arbitrary set of triangle meshes together in a single mesh.
Convert to point cloud Creates an unstructured point cloud from the vertices of the mesh using
the color attribute of the mesh
Get mesh borders as polyline To create a new polyline containing the mesh's borders and add it to the
project
Mesh point list Enables you to select points for a mesh.

Mesh Editor
This dialog is an environment designed for advanced mesh editing operations. It works on one mesh at a time. It is
accessible from by selecting Tools > Meshing > Mesh editor from the corresponding button in the top toolbar, or from
the context menu of any mesh in the Project Window.
From top to bottom, the dialog displays the mesh's full path, then a toolbar with groups of buttons, then a navigable 3D
view to show the mesh, and lastly information on then number of triangles and vertices present.
File
This button group has the Save and Save as options. These are needed because the edits performed in this dialog are not
automatically transferred to the mesh in the ATLAScan project. If you close the dialog you are asked if you want to save
or discard the changes.
Geometry
This button group allows you to switch between the three visualization possibilities of a mesh: Vertices only, triangle
edges only (wireframe), triangle faces (flat). The button backface allows you to show or hide the triangles' backface.
Coloring
The first two buttons here allow you to change how to color the mesh: Using the colors associated to the vertices, or
using a gray color that changes only according to the triangles' inclination with respect to the light source. The
Boundaries button highlights in bright red all the boundary edges of the mesh, which is particularly useful for quickly
spotting where the mesh has holes to fill or imperfect boundaries that should be corrected.

Selection
There are several selection modes. When the button None is checked, you can navigate the mesh. By clicking Triangles,
you start selecting the mesh triangles by drawing a rectangle on the viewport. Selected triangles are highlighted in
yellow. The button Invert enables you to invert the current triangles selection. When you are drawing the selection
rectangle, keeping Shift pressed will enable you to add triangles to the current selection by drawing another selection
rectangle. Clicking Boundary lets you select one of the circular boundaries that the mesh has. Boundaries are polygons
formed by those mesh edges that belong to one and only one triangle. A closed mesh has no boundaries. Boundaries
selected in this mode may delimit a hole that you want to close. After selecting the hole's boundaries, you can click on
Close boundaries in the Editing button group.
Editing
Once you have selected some triangles with the Triangles button in the SAelection group, you can delete the selected
triangles with Delete triangles. Clicking Close boundaries closes a mesh hole whose boundaries have been selected. This
function works best with small and close-to-planar holes.
Filter
This button group contains three functions:
Decimation
This tool allows you to simplify your mesh by substituting small triangles with bigger ones if the mean-square error
introduced in this way is smaller than a user-customizable threshold. Simplification is based on conveniently collapsing
edges among triangles.
For each edge shared by two triangles, ATLAScan assesses whether to collapse it or not. Collapsing an edge means
assigning the edge's two vertices to coincide in the edge's middle point, and adjusting all neighboring triangles
accordingly. If the quadratic error between the surface before and after the collapsing is below the parameter Max
quadric error, then the edge is collapsed.
The algorithm tries to collapse a percentage of edges as specified by the parameter Edge collapses. For example, if it is
set to 0.1, ATLAScan will try to collapse 10% of the edges.

Denoise
This algorithm allows you to de-noise your mesh, meaning to iteratively eliminate spikes from your mesh, without
changing the surface's volume.
A smoothing filter is run on the mesh for the given number of Iterations. This filter has anisotropic properties: it
eliminates the spikes without changing the surface's volume. During the iterations, it distributes the volume from the
spike to the neighboring triangles.
Ridges & Valleys
This dialog could also be called crests & toes. It is a semi-automatic technique to detect ridges (prominent mesh edges)
and valleys (reentrant mesh edges). When you open this dialog, the ridges and valleys are calculated and drawn in the

mesh editor. Ridges are drawn as blue polylines, valleys in red. Ridges and valleys can be then saved as polylines.
The idea is that tool shows you an initial extraction of ridges and valleys, and you successively refine the extraction
parameters and see the effects of your changes in real time. You can refine the parameters until you are satisfied, then
save the ridges and valleys as polylines in the project. The parameters regard the curvature, the horizontality, the length
of the edges and the gaps between them. While you modify the parameters and recompute the ridges and valleys, you
can choose to turn on or off the visualization of the ridges, the valleys, and the mesh via the three check-boxes in the
Draw group at the bottom of the dialog.
Curvature filter
The topmost slider of the dialog allows you to modify a threshold on the curvature of the ridges and valleys. Ridges and
valleys are edges shared by your mesh's triangles. Each edge has an associated curvature, depending on the angle
between the two associated triangles. A threshold on the curvature filters out the smoother edges and leaves only the
steeper ones, which normally are the most important. Scroll the slider to see the effect of the threshold in real time.
Orientation filter
Below the curvature filter, you can also activate a orientation filter. This filter keeps only the edges whose angle with
respect to the horizontal plane is close to the angle you indicate with the slider. In this way, you can choose to keep only
the horizontal edges, or the vertical ones, or the ones with a given inclination.
Close gaps in edges

Sometimes ridges and valleys come in many segments with small gaps among them. In the groupbox Close gaps in
edges you find a tool to close those gaps and get the ridges and valleys as continuous polylines. You can set the
maximum gap distance to close, and press Apply. ATLAScan will start closing the gaps from the smallest to the biggest,
without creating loops or bifurcations.
Delete short edges

Delete short edges contains a tool to filter out polylines whose total length is less that the specified parameter. Normally
you want to discard short edges because they are not the main ones. The main ones should instead be found by closing
the gaps among them.
General procedure
As general procedure, it is recommended that you use the described filters in the order given above, and close and
reopen the dialog if the result is not satisfactory. Once the extracted ridges and valleys are good enough to describe the
important edges of your model, press Save as polylines at the bottom of the dialog. New polylines with the ridges and
valleys will be added to your project.

Mesh selection from current view point
This function allows to cut a portion of a mesh using 2D video selection tools on the current view. It is accessible through
Tools > Meshing > Mesh selection from current view point. When you activate this function, the window above appears
in the recipe window, and ATLAScan's 3D window goes into Selection mode.
Drag a mesh from the project window and drop it on the window shown above. The mesh must be loaded for the
procedure to work. Use one of the three video selection tools to select the desired portion of mesh, and click Select. The
dialog shown below appears:
The dialog above shows the name of the mesh created from the selection, and information about area and number of
triangles. After closing the dialog, you may want to go back to Navigation mode by, for example, pressing the space bar.

Mesh selection with 3D polyline
The 3D mesh selection works with an input mesh, a 3D polyline and a viewpoint that can be defined by an orthocamera,
by a plane or by the current view.
The polyline, seen by the input viewpoint, determines a frustum that intercepts the mesh defining a portion of it. This
portion is acquired by doing Virtual scan, and uniformly meshing the resulting grid point cloud. The virtual scan is always
done with an orthocamera that ATLAScan creates internally.
When selecting the polyline and the viewpoint, the user doesn't need to check whether the polyline is contained in the
orthocamera's frustum, or how the polyline is positioned with respect to the plane. In fact, ATLAScan creates an
orthocamera whose frustum is automatically enlarged to include all the polyline. This automatic expansion, however,
does not take into account the mesh (this is done to fully exploit the orthocamera's resolution on the selection region).
The resolution of the internal orthocamera is calculated proportionally to the number of vertices in the input mesh, so
that the scanning is precise enough to preserve the input mesh's features.
If a plane is given as input, ATLAScan creates an orthocamera that points in the plane's +Z direction. If the polyline and
the mesh are in the -Z semispace, they will be included anyway in the selection because the orthocamera will have a
negative near plane. However, the direction of the plane (or of the orthocamera) influences the winding direction of the
mesh that results from the selection.

Make single mesh
This dialog allows to lump together in a single mesh an arbitrary set of triangle meshes. To activate this dialog, you can
select a set of triangle meshes from the project window and select Meshing > Make single mesh from the contextual
menu or from the Tools menu.
To add more meshes to cluster together, click Add files at the top left. Clicking Process starts the clustering. A new mesh
is created that contains just the triangles of the input meshes.
See also clustering tools and triangle meshes

Photo & Color

ATLAScan provides several coloring tools to effectively manage the colorization of point clouds and meshes. These tools
can be classified according to the objects you want to color and the procedures you need to apply:
Some color information is item intrinsic of point clouds and saved as a color layer:
Histogram Lets you optimize the point cloud's color contrast stretching the
histogram.
Colors mapping Creates an artificial colorization for a given color layer of a point cloud.
Color information can be scalar (e.g. reflectance, range, confidence) or
vectorial (e.g. inclination). It’s also possible to add layers from external
images.
Color with altitudes Adds an extra color layer to the selected point cloud(s), representing the
altitude of the points with respect to one of the tree axes of the
current UCS.
Inclination from plane Adds an extra color layer to the point selected cloud(s), representing the
inclination of the points' normals with respect to a given plane that exists
in the project. Suggestion: You can use this to classify points for further
vegetation removal.
A mapping (on point clouds and meshes) of an image from external cameras is possible:
Color clouds with photos To create a color layer for a structured point cloud by blending the
images from selected projectors.
Create Projector To create a Perspective, Orthographic, Spherical, Cylindrical projector or
from a calibration (by importing a ATLAScan camera calibration file) using
a valid image.
Camera calibration on a grid To register photos with the 3D data (grid point cloud). This process
creates photo-realistic texture maps for the 3D model.
Camera calibration using 3D To calibrate a perspective image using a list of matches between camera
points points and 3D world coordinates points.
Calibrate camera intrinsics To calibrate perspective cameras intrinsic (or internal) parameters using
multiple checkerboard images as calibration samples.
Camera Creates a panorama from models and images.

Import georeferenced TIFF This function imports a geo-referenced TIFF as an orthographic camera.
Create texture map To create textures for mesh models.
The coloring tools are accessible from the Tools > Photo & Color menu or from the Photo & Color item in the point
clouds context menu. See also Photo and Color in the Top Toolbar. To look at point clouds’ color layers, select the point
cloud from the project window and set the Color Mapped option in the Property Browser.
See also our video tutorials on point clouds coloring.

Histogram dialog
The histogram shown in the dialog represents how many points belong to a given color value. This tool is available only if
the cloud is loaded in memory and if the current color type is scalar (it doesn't work on Inclination for example). Given a
point cloud and one of its scalar color layers (e.g. reflectance), the histogram tool allows you to optimize the color's
contrast by stretching the histogram.
• The two sliders allow you to decide the start and end of the histogram interval to be rendered.
• The Auto Cut-off button allows you to automatically set the histogram's minimum and maximum bins so that a
certain percentage (2% in the figure above) of the histogram energy is discarded starting from zero to the minimum
value and starting from the highest histogram bin back to the maximum value.
See also Point clouds and Colors mapping dialog.

Colors mapping
The purpose of this command is to manage and to create an artificial colorization for a given color layer of
a point cloud. The command is accessible from the Photo & Color menu in both on the Tools Menu bar and in
the Top toolbar.
You can still use ATLAScan's main GUI while this dialog is open.
When a point cloud is pre-processed, extra information is added, which is organized into color layers. Information can be
scalar (e.g. reflectance, range, error bound, confidence) or vectorial (e.g. inclination). You can add color layers to a cloud
via functions like Inspection or Color with altitudes, or by clicking the Add color layer button in this window.
Only one color layer at a time is rendered in the 3D scene. You can change a cloud's current color from the layers list on
the left of the colors mapping dialog or by clicking on property browser. The change takes place after pressing
the Apply button.
When a color layer is rendered, a colorization is used to associate the points' color values to a drawable color. Only
scalar information can be mapped with colors through this dialog. The color layer Inclination is an exception: it
represents the normals of the points and it is colored automatically.

Map colors tab
Layers list
On the left there is a layers list that shows which layers are associated with the current point cloud.
Predefined colors buttons

In the center there is a set of buttons to apply a pre-defined colorization to the currently selected layer. Some are
discrete while some are continuous.

The buttons used to map colors are:

• Measure values (continuous): Map the current layer to a continuous Rainbow Matlab style color.
• Measure values: Map the current layer to a discrete Rainbow Matlab style color.
• Single thresholding: Map the current layer to solid red and blue. You can set a threshold moving up and down the
central value label.
• Double thresholding: Map the current layer to solid red and blue. You can set two thresholds moving up and down
the values labels.
• Gray scale: Map the current layer to a continuous gray gradient.
• Gray scale (discrete): Map the current layer to a discrete gray gradient.
Color scale
On the right there is a color scale that shows the selected colors map in combination with labels. Labels show maximum
and minimum (bounding) values and also values between colors. You can choose among several types of color scales via
the Predefined buttons or by clicking on the Advanced button. You can also disable colors by clicking on the color check
box.
Advanced controls
The user can master colors and design or produce to meet his or someone's individual requirements.
Advanced commands are divided in three groups.
Legend properties:
• Max: Sets the max value and shows it in the legend
• Min: Sets the min value and shows it in the legend
• Colors spacing: Gives information about width of the color
• Number of colors: Set the number of colors to use
• Use gradient: Switch between a discrete and a continuous colorization
• Decimals: Set how many decimal digits to use
Thresholding group:
• Keep points above max: Keeps points above "Max" value else points are hidden.
• Keep point below max: Keeps points below "Min" value else points are hidden.
• Keep point of disabled intervals: Keeps points in the deselected color else points are hidden.
• Minimum confidence: Manages confidence value.
Color group:
• The combo box lets you chose between the following color styles:
1. Single thresholding
2. Double thresholding
3. Extended
4. Grayscale
5. Rainbow
6. Rainbow Matlab
7. Ironbow
8. Redblue
9. Customize
• Use default color when out of range: Applies the default color (gray color determined by reflectance value) if the
main color map would be out of range.

Histogram tab
A colored histogram shows a statistical distribution with colors chosen by user. In the abscissa, the histogram
information about current measure (each color identifies a local radius of the measure) and in the ordinate it shows a
measure of how many occurrences of points are mapped to a color. The mean value is also shown.

Add color layer tab
You can add customized color layers by clicking Add color layer tab and selecting one of the following:
• Color with altitudes: Inserts a layer to color distances along an axis direction.
• Inspection: Inserts a layer to color distances between point cloud and a reference model (e.g. Mesh) .
• Compute point normals: Inserts a layer to colors normals of points coherent to a plane direction.
• Compute confidence: Inserts a layer to color confidence amplitudes.
• Inclination from plane: Inserts a layer to color inclination angles.
• Color with range: Insert a layer to color distance from the observation point.
• Merge or split layers: Combines RGB and scalar layers to obtain a new custom color layer.
• Add layer from image: Applies an image on the point cloud (e.g. a 2Dview after a color editing). Pay attention to the
dimensions of the images: they must be the same of the point cloud ones.

Color with altitude

This tool works on any set of point clouds. It adds an extra color layer representing the altitude of the points with respect
to one of the tree axes of the current UCS to the selected cloud(s).

Inclination from plane

This tool works on any set of point clouds. It allows you to add an extra color layer, representing the inclination of the
points' normals with respect to a given plane that exists in the project, to the selected cloud(s). The inclination from the
plane ranges from -1 to 1 and is calculated as the scalar product between the points' normals and the plane's normal.
As an example, it is possible to use this command to distinguish points on vegetation from points on soil (with regular
inclination), making it useful for vegetation removal.

Color clouds with photos

Projectors blending dialog
This dialog allows you to create a color layer for a structured point cloud by blending the images from selected
projectors. You can load all the projectors available in the current project. The color layer is computed by projecting on
the point cloud and blending all the images coming from the projectors. Different blending algorithms are applied
depending on the selected options.
• Blending Result: Shows the image resulting from the blending process.
• Projectors to blend: Selects the projectors to be blended by checking the items in the list. You must ensure that the
projectors are properly calibrated for the selected point cloud, otherwise the correctness of result will not be
guaranteed.
• Apply as color layer: Adds the blending result to the structured point cloud as a new color layer. Upon clicking the
button you will be prompted for the name of the layer.
• Save as image: Saves the current blending result as an image. The image can be loaded at a later stage as a color
layer (see Colors Mapping for further information on this subject).
Blending options
If Colors missing points with reflectance is checked, points that cannot be colored using projectors are colored with
their reflectance value. If this option is left unchecked, the missing points are blacked out.
See also Texture mapping dialog and Camera calibration dialog.

Create Projector
This functions creates a projector. Five different modes are available:
• From calibration: Creates a projector (perspective or spherical) by importing a camera calibration file (CAL).
• Perspective: Creates a perspective projector.
• Orthographic: Creates an orthographic projector.
• Spherical: Creates a spherical projector.
• Cylindrical: Creates a cylindrical projector.
Note that a valid image is needed in order to create a projector.
See also Cameras, and Camera Calibration on a Grid.

Edit (projector) image dialog

This dialog is available only for projectors. Here is possible to edit the transparent parts of the image in order to optimize
the blending of projectors. The editing dialog of an image of a projector is accessible through the contextual menu of the
loaded image (Edit projector image).

Camera Calibration on a grid

The objective of camera calibration is the registration of photos (perspective or spherical images) with the 3D
data. This process creates photo-realistic texture maps for the 3D model.
When you click Photo & Color > Camera calibration on a grid from the Tools menu, from the Tools item in the toolbar,
or from a point cloud's context menu, the Camera calibration dialog appears.
Load a photo to calibrate with Load image, and at least a grid point cloud with Load 3D grid. If necessary, change the
reference grid from the loaded ones; the image-grid point pairs are stored for each grid, and are accumulated.
A resizable zoom window appears for both where an independent zoom factor (Alt+mouse wheel) can be set from the
main window; simply center the zoom area by moving the mouse on the main window while pressing Ctrl. To place a
marker, double-click in either the main window or the zoom window. To move a marker, left-click on it and drag it to the
new location. To remove the marker, right-click on it.
In general, to interact with the images:

• Zoom in/out: mouse wheel
• Fit the image in the window: Alt + middle mouse button
• Reset to original size: Alt + right mouse button
• Translate the image: Shift + drag with left mouse button.
• Rotate image 90°: Space bar.
• Show/hide zoom window: Z key
• Move zoom region inside global view: Ctrl + left mouse button
The list of points is shown in the table, where every cell is manually editable to force the values. It is possible to import a
list of points from a file by clicking Load points.
For the photo, the format must be

Image_x1 Image_y1 <Point1 Name>
Image_x2 Image_y2 <Point2 Name>
...
For the grid the format must be

X1 Y1 Z1 Image_x1 Image_y1 <Point1 Name>
X2 Y2 Z2 Image_x2 Image_y2 <Point2 Name>
...
Try to change the color type of the grid on top of the window to improve the contrast of features. Press the right arrow
to show a menu of available commands for the current color type for better adjustment.
Choose the right calibration algorithm, depending on your kind of photos:

• OpenCV for perspective images
• Spherical Camera for spherical images
After you have found a sufficient number of marker pairs between the photo and the grid (at least 11 pairs for
perspective images and at least 4 pairs for spherical images), click Calibrate. The result is shown in the Log panel with the
computed field of view of the photo and relative re-projection error. If successful, the calibration and the point pairs can
be saved to file by pressing Save calibration. A CAL file is created in XML format storing the calibration using calibration
parameters. As a test of the calibration, click Make Reprojection, to save an image of the photo re-projected on the grid.

Point guessing system

When you select at least 11 correspondences for perspective images (and 4 correspondences for spherical images), the
point guessing system (PGS) will perform a background calibration and will automatically try to estimate the position on
the image of additional points that you place on the grid.
The procedure is the following:

• Select a point on the grid (right image).
• The PGS will place the guessed point on the image (on the left). The zoom box will center on the newly placed
point.
• If the PGS puts the point in the wrong place in the image, correct it.
• If the PGS cannot place the point in the image at all, place it manually.
Note: You can also select a point on the image and the PGS will try to place its corresponding point on the grid, but this
procedure is not recommended because it is slower and less accurate.

Camera Calibration Using 3D Points
This dialog allows the calibration of a perspective image using a list of matches between camera points and 3D world
coordinates points. First, the image to be calibrated must be selected by using the Load Image button. Subsequently
image points and 3D points must be loaded using the Load Image Points and Load 3D Points buttons respectively.
Please refer to the camera calibration page for details about the calibration options.

Calibrate Camera Intrinsics

If you are using an external perspective camera (not connected to the laser scanner system) to calibrate high definition
photos on grid point clouds, it could be necessary to create a specific calibration for the camera, save it, and use it for
future purposes.
This page shows the procedure to calibrate perspective camera's intrinsic (or internal) parameters using multiple
checkerboard images as calibration samples.
This tool is accessible by clicking Tools > Photo & Color > Calibrate camera intrinsics in the menu or in the
top Photo & Color toolbar.
1. Print the calibration frame.

2. Fixed it on a rigid support.
3. Measure the high and length of the square.
4. Select the camera and the lens you want to calibrate.
5. Acquire at least 10 images (to have a robust estimate for the calibration parameters) of the frame. Change
the camera position between each image, and make sure to capture all the frame squares in each image so
that you can reduce the image without the frame and to avoid glare effects.

6. Create a new ATLAScan project and select Tools > Photo & Color > Calibrate camera intrinsics
7. Set the square size and the number of internal corners and load the frames images.
 Chessboard square size: The width and height of a single chessboard square.
 Chessboard internal corners: The number of corners of the chessboard. Note: The border corners should not
be taken into account.
 Camera sensor size: The width and height of the camera sensor. Refer to your camera's technical
documentations to obtain this data.
In this sample image, a chessboard with 9 by 6 internal corners is shown. Please note that, in order for the algorithm to work properly,
the number of row and column corners must be different.
8. The calibration process starts automatically

9. At the end of the calibration process, check the mean re-projection error to verify the process quality.
10. Click Save intrinsic to store the calibration file (Internal-Calibration.cal), which is usable for all the other
pictures.
Once the calibration is complete, a calibration file can be saved. Such file can be used as a starting point for camera
parameters while using the camera calibration function. Inside the Camera calibration on a grid dialog, select External
and click Load internal calibration to select the file you just created.
Notes
• After selecting the camera and fixing the focal length, the internal calibration can be done once for
several acquisition jobs.
• After intense usage of the camera, we suggest you to verify the calibration parameters.

Panorama creator
This tool is accessible by clicking Tools > Photo & Color > Camera in the menu or in the top Photo & Color toolbar.

Import Georeferenced TIFF

This function imports a geo-referenced TIFF as an orthographic camera.
The image to be imported must be accompained by a Tiff World File (TFW) indicating the coordinates of the image.
See also projectors and UAV and imaging.

Texture mapping
The command Create texture map allows you to create textures for your mesh model.
You can simultaneously load up to four projectors that project photographic color on the same mesh. Then, you must
create a camera to specify a viewpoint from which the texture should be created and parametrized. Finally, you can
create a texture.
A texture is a particular image that, mapped on your mesh, gives it its color. The texture is calculated by blending
together all the color coming from the projectors.
Preview resolution:
• Update preview: Creates a preview of the texture from the selected view point and projection. This is a
simplified image and does not represent the full quality of the final texture
Projectors to blend:
• Create single texture map: Creates the texture by blending the checked projectors. A PNG image is created,
plus a 2f file that stores the computed texture coordinates for the mesh vertexes, which are listed in
Mappings. Also, you can select whether or not to Compute shadows.
Mappings:
• Remove: Removes the selected mapping from the list.
• Remove all: Clears the list.
Re-cluster: Creates a new triangle meshes model where the listed mappings are used to re-arrange the triangles in a new
set of submeshes that map the desired textures. Each triangle is mapped with the best texture quality available by
computing the best texture projection among all of them. The model is automatically added to the project.

Windows menu
All the dockable windows can be opened by right-clicking on the top toolbar or by selecting them through the Windows
menu item.
• Project window
• Property window
• Readout window
• Log window
• Manual positioning (Adjust Pose)
• Info window
• Recipe window
• GUI options: Opens a dialog to change the graphic user interface (GUI) settings.
• Readout window

Project Window
This window provides a tree view on the current project items.
The Project window lists the items loaded in the project, organized in a tree structure, where each data type is grouped
in a separate folder. By hovering the mouse over the item name, a tool tip pop ups with some information about the
item, like file path, file size, etc.
To enable the rendering of an item, click the checkbox on the left of the item. The checkbox of a folder enables/disables
all its sub-items.
Each item can be manipulated by right-clicking on it, which opens a context menu that display several commands. Each
item has properties that can be viewed and edited in the Properties window when the item is highlighted.
Items such as point clouds or triangle meshes are models. Models can be loaded or unloaded, loaded models are listed
in bold in the project window. To rename a model, right-click on it and select Rename. To rename all other
items, double-click on the item's name and edit the name.
To remove an item from the current project, select the item in the Project Window and press the Delete key on your
keyboard. The files are not lost, they are simply moved to the project's Trash folder inside the project directory, so they
can be recovered.

It is possible to select multiple items in the tree using Shift and Ctrl keys. You can then:
• Delete the selection with Delete key
• Set common properties in the property window
• Open the context menu by right-clicking on the selection to display the commands that are applicable to the
selection.
Quick model distinction can be made by clicking Color by ID on the Project Window's toolbar. Each model is colored by
its color ID set in the Properties window. This color is randomly computed when the model is inserted in the project.
See also dockable windows, and project items.

Recently used items dialog
This dialog lets you quickly select some of the project items you used most recently.
Many important ATLAScan functions require project items as input. For example, the Registration function needs
two grid point clouds as input. You can start these functions by selecting the input items first, in the project window.
However, this dialog offers you an alternative and in some cases more efficient way of starting your functions.
If you start a function that needs items of a certain type as input, and none of these items is selected, then the dialog
pictured above pops up to help you. An important feature of this dialog is that it shows you the candidate items for your
function, listing them from the most recently used to the least recently used. If you are working repeatedly around few
items, using this dialog might be more efficient than scrolling through the project window each time, especially if it
contains hundreds of items.
See also project items, and project window.

Property Browser Window

This dockable window is situated by default on the left side of ATLAScan's GUI. It allows to read and edit
the properties of the current selected item(s) in the project window.
In ATLAScan, each project item has properties. There are general properties, owned by all project items, and there
are specific properties that are only apply to certain types types of item.
General properties
These properties are common to all ATLAScan project items, and always displayed in the property editor:
• Hyperlink: Each project item can be linked to any URL, other project items, files on the PC, and web
addresses.
• Color ID: Each project item has a specific color to identify it from other items. You can edit the color ID by
clicking on the value field of the color ID property. To render the items with their color IDs, select Navigation
> Color by ID.
• Comment: This comment will appear in the item's annotation if the above property is true.
• Draw annotation: If this boolean property is true, each project item's annotation, is rendered in the 3D scene.
• Draw coordinate frame: This property is not owned by annotations. All other project items have an object
coordinate system. If this boolean property is true, ATLAScan will render an axes triplet to show where the
object coordinate system is. This is useful for tasks like finding out the direction of a plane's normal.

Information window
The Information Window allows you to see detailed information about the object you have selected in the Project
Window and perform certain actions on it. The exact information displayed depends on the type of object selected. For
example, a mesh object shows information about its borders, while a point cloud shows information about the points it
contains. The Information window is accessible by right-clicking on the top toolbar.

Readout window
The Readout Window allows you to see the 3D coordinates of the points you are hovering the mouse on in the 3D
rendering window. It is accessible by right-clicking on the top toolbar.
The readout window displays information about the 3D point that the mouse is currently hovering over:
• Coordinate System
• Global Coordinates System: The 3D point's X, Y and Z coordinates
• Current User Coordinates System: The 3D point's X,Y and Z coordinates, expressed in the current UCS.
• Pixel column, row: The mouse's position inside the 3D window, in pixels (0, 0 is the top-left corner)
• Pixel color RGB: The point's color, expressed in its red, green and blue components.
• Depth to point: The distance between the 3D point and current view point
Clicking Change units of measure opens the Select Units of Measure window, which lets you change the displayed units
of measures (SI, imperial or U.S. customary units).
The readout window appears docked by default in the right area of the GUI.

Log Window
The Log Window records everything that happens in ATLAScan, including button clicks, algorithm status updates, errors,
warnings, procedure results and more. Log messages can be of four types: Information, warning, critical and fatal. Log
messages appear always with their type and timestamp.
The log window is docked in the bottom right area of the GUI by default.

GUI Options dialog

This dialog allows the user to change settings for the graphical user interface (GUI).
The Dock widgets can be moved and Dock widgets can be detached as separate windows checkboxes control how
freely the dockable widgets can be moved around and detached.
The Reset dock widgets position button resets the position of the widgets to their defaults. These defaults are decided
by the current workspace configuration, internally known by ATLAScan.

Custom toolbar editor

In addition to the preset workspaces available in the Workspace menu, there is also the Custom Workspace. The Custom
Toolbar Editor lets you select which buttons are shown in the custom workspace's top toolbar.
You can set up multiple profiles with different sets of tools for different purposes. For instance, you could create one
profile for making measurements, another for performing registration, and so on.
To create a new profile, click New profile and enter a name for it. To edit an existing profile, select it from the Select
toolbar profile drop-down menu.
The central panel is a tree containing all of the possible buttons that could be shown in the toolbar. To add a button to
the selected profile, simply drag the desired button from the tree and into the Toolbar Builder panel. To remove a
button from the selected profile, drag it from the Toolbar Builder back into the tree.
Before you can use a profile in the Custom Workspace, you must set set it as the active profile. To do this, select it from
the drop-down menu and click Set as active profile. The next time you select the Custom Workspace, it will show the
tools in this profile.

Workspace
This menu item contains a list of workspaces. Each workspace has a different set of tools available in the top toolbar and
in the menu bar. Some workspaces are only available if you have a particular type of license for ATLAScan.
To alter the buttons displayed for the Custom Workspace, open the Custom Toolbar Editor by selecting Windows >
Toolbar Manager from the menu.

Help
Through this menu you can access ATLAScan's Online Help system and get diagnostic information, which is useful when
asking for support. Moreover, you can enter ATLAScan's license manger to browse the license features activated and
activate new ones.
• Online help
• About plugins
• About ATLAScan

Online help
This menu voice opens the present online help. You can also open the online help by pressing F1 (if you do so while in a
particular window, the help system will open to that window's help page).

About plugins
The menu option About plugins shows you a list of plugins. This is useful to know which import/export formats are
currently available in ATLAScan, and to know if a particular algorithm (e.g. LM-ICP bundle adjustment) is installed or not.
When asking for support, you may be asked to send a screenshot of this window.

About ATLAScan
This menu option opens the About ATLAScan window. This window displays ATLAScan's complete version number and
other diagnostic information. When asking for support from Teledyne Optech, you may be asked to produce a
screenshot of this dialog, or to report some information from this window.

General Concepts
In this section you can find the main concepts that you can encounter when using ATLAScan.

Annotations
ATLAScan allows you to place annotations of your 3D models in the scene. These annotations may appear as items in
your project. Annotations can be of three types:
• Point annotations
• Distances
• Angles
You can create an annotation by selecting Output > Measures & Notes > Annotation in the menu or the corresponding
buttons in the top toolbar. This opens the annotation recipe window. To create a point annotation, click Annotation and
double-click on the desired point in the 3D scene.
Distance annotations are used to measure the distance between two points in the 3D scene. To create a distance
annotation, select Output > Measures & Notes > Distance from the menu bar, then left-click on the first point you want
to measure, drag the mouse to the second point, and release the mouse. ATLAScan shows you the distance measured
and asks you whether you want to store it in the project.
To measure angles, select Output > Measures & Notes > Angle in the menu bar. You define the angle you want to
measure by left-clicking on three points of your 3D scene.
In the properties of any annotation, you can change its color ID, edit the displayed comment, define a hyperlink for an
annotation (which lets you add meta information to your 3D scene from other places, such as the web) and choose
whether to draw the annotation or not.
Annotations are stored in the form of point collections. Therefore, they don't have a pose matrix like all the other project
entities and cannot be moved or rotated through the Adjust Pose tool or the Pose dialog.

Cameras
A camera is a project entity that defines a viewpoint in your virtual 3D world. A camera has a pose (like all project
entities) that defines the camera's position and orientation. A camera has also a projection that defines how the camera
sees the world.
There are four types of cameras, according to their different projections:
• Perspective cameras
• Orthographic cameras
• Cylindrical cameras
• Spherical cameras
The editable properties of the cameras are:

• Comment: User-editable comment
• Draw coordinate frame: Draws the local coordinate frame of the frustum
• Clip far plane: Distance of far clipping plane in meters. Specifies the maximum rendered depth. Must be
greater than Min depth.
• Clip near plane: Distance of near clipping plane in meters. Specifies the minimum rendered depth. For
perspective cameras, only values greater than zero are allowed
• Image size (Width, Height): Size of the image in pixels (used only as a place holder for repetitive virtual
scanning)
• Shadow depth bias: Adjusts the shadow calculation for a projector. Sometimes, during shadow calculation,
there can be some noise while assessing which object is in foreground and which is in shadow. If you increase
this parameter, the effect of noise is reduced, however the shadows risk becoming unrealistic. If you decrease
it, you will get more precise shadows, but you risk getting a "salt and pepper" effect, as if the projected color
also fell on the object in shadow in small grains.
Furthermore, there are specific properties for each of the camera projection types:
• Perspective cameras:
o Vertical field of view (degrees)
• Orthographic cameras:
o Extent on X [m]
o Extent on Y [m]
o Keep aspect ratio for image size
• Cylindrical cameras:
o Slices
o Length [m]
o Radius [m]
o Longitude begin [deg]
o Longitude end [deg]
• Spherical cameras:
o Slices
o Stacks
o Radius [m]
o Longitude begin [deg]
o Longitude end [deg]
o Latitude begin [deg]
o Latitude end [deg]

These particular properties, which change with the camera type, define the frustum of the camera, together with the
near and far plane. The frustum of a camera is the 3D region of space that the camera sees. It is a pyramid for
perspective cameras, a 3D rectangle for orthocameras, a cylinder for cylindrical cameras, and a sphere for spherical
cameras.
From the contextual menu of any camera, you can also activate the following commands:
• Go to: Jumps to the camera view point.
• Invert direction: Inverts the direction of the view of the camera
• Apply projection: Jumps to the camera view point and applies the camera projection as the current rendering
projection (if perspective or ortho)
• Duplicate as: Creates a new camera copying the transformation of the current one
• Fitting
o Set optimal depth range: Automatically computes the tightest near and far clip plane positions
o Select projectors in this frustum: Controls which cameras have both a) a focal axis that forms an
angle with the focal axis of the selected chamber lower than Maximum incident angle, and b) have a frustum
that intersects the frustum of the selected camera. The projectors/cameras are checked in the project
window.
o Select models in this frustum: Finds and checks the meshes intersecting the selected camera (this
is useful for texture mapping).
• Elevation&Plan > Virtual scan: Uses the camera frustum to resample the scene and generate a new grid point
cloud or an image.
Warning: To optimize the depth accuracy of the virtual scanner, try to keep the clip near and far planes as close as
possible to the desired scene depth range.

Flythroughs (video trajectories)

A flythrough, or video trajectory, is a trajectory that runs in your 3D scene, defining how a video camera will move
around in your scene to capture a movie. A flythrough is the “backbone” of the movie you're creating of your 3D scene.
In the picture below, the red curve line is the trajectory that the video camera travels. The yellow segments are the
optical axes of the video camera for each video frame.
Flythrough creation
The easiest way to create a flythrough is by using the flythrough editor: you can define keys, or important positions you
want your video to pass through, and then interpolate the remaining frames. You can also create a flythrough via the
point list window. The difference is that with the point list window you can only use points that belong to existing
models to define your trajectory, while with the flythrough editor lets you choose your keys in total freedom.
Operations on flythroughs
If you right-click on any flythrough, you have access to several useful operations. You can play the flythrough, to see a
preview of the video associated with it. You can edit the flythrough, which causes the flythrough editor to appear to
allow you to add or delete keys to your trajectory. You can make the final movie product by specifying the video
encoding parameters using the movie dialog. These three options are also available through the top menu, via Outputs >
Video record.

From the flythrough contextual menu you can access the Tunnel survey submenu, which has two functions designed for
flythroughs:
• Cylinder virtual scan. This function will enable you to virtually scan your 3D scene by looking at it from a cylindrical
camera whose “backbone” is the flythrough. More specifically, ATLAScan will split your trajectory in many segments,
each segment running from point T(t) to point T(t + 1), where T(t) indicates the point on trajectory T at time t. Out of
each one of those segments, a cylindrical camera will be created having as main axis the segment. ATLAScan will pop
up as many virtual scan dialogs as there are segments, to allow you to do a virtual scan.
• Generate cross sections. This function provides you with a way to create cross sections of your models along the
trajectory. If you select this function, ATLAScan asks you the spacing between different cross sections along the
trajectory. When the spacing is set, ATLAScan creates several planes, with the normal placed along the trajectory
according to the spacing selected. These planes are added to the project and named Flythrough section n(N). It is
left to you to actually select one of these planes and calculate the desired cross section.
You can also convert the flythroughs into polylines by using the Convert to polyline command in the contextual menu.

Groups
A group is a cluster of items composing the project. You can create and manage the groups from the contextual menu.
To create a group, right-click on a data type folder to open its context menu and select New group.
You can populate a group in one of the following ways:

• Move a selection of items from their previous position to the new one with Move command from the
contextual menu of the selection
• Paste the selection of items in the new group, with Paste in group command from the contextual menu of
the new group.
In the group's Property Browser you can add a Hyperlink and set the Registration role of all the scans inside the group:
• None
• Same as parent: Inherits the registration role from the higher level
• Children move together: If a single scan is moved in the 3D space, all the other scans move together, as a
rigid system
These roles are used to make a scans registration through groups.
You can also create a group made up of different data type objects, as shown below. This is is useful for putting together
all the items of a single sub-project, for example.

To remove an object from a group, select Ungroup from the context menu.

Planes
ATLAScan allows you to define planes in your project.
A plane has a width and a height that can be specified in the property editor window under Extent on X and Extent on Y.
A plane has also all the properties of a project entity: Color ID, Comment, Hyperlink, Draw annotation, and Draw
coordinate frame. To enable the last one is particularly useful in order to visualize the plane's normal direction.
You can select the following commands from a plane's context menu:
• Cross section: Computes a cross section of the models activated in the scene. The result is a polyline.
• Invert direction: Inverts the plane's coordinate frame Z axis direction. The positive direction determined by the
plane influences certain other functions, such as cross sections, volume, and cut and fill calculations.
• Meshing > Create mesh from a plane: Creates a mesh starting from the plane object. Useful for example for
inspection tool.
• Set constrained draw: Picks points in the 3D scene and draws them as a projection on the plane.
See also Project items and Plane creator/editor.

Point Clouds
There are two different kinds of point clouds in ATLAScan:
• Structured point clouds (also called grid point clouds, range images, or scans): These are scan-based clouds,
"fixed-origin" or tripod scans. They include row/column info and the origin scan’s position.
• Unstructured point clouds.
The following operations are available for both unstructured and grid point clouds:
• Recompute bounding box: This is useful if the bounding box is not updated (available only if the model is
loaded in memory)
• Histogram: Lets you to optimize the contrast stretching the histogram (available only if the model is loaded in
memory and if the current color type is 1f or 1d, i.e. a high-dynamic single-channel point color)
• Colors mapping: Remaps point colors to a pseudo colored scale to improve the dynamic range. This is useful
if the histogram stretching is not sufficient (available only if the model is loaded in memory and if the current
color type is 1f or 1d, i.e. a high-dynamic single-channel point color). This tool can be also used for range
segmentation and inspection.
• Inclination from plane: If the Inclination color type is available (for which the model must be pre-processed),
this dialog allows you to select a plane whose normal is projected to the normals of the model (dot product).
The result is added as a 1f color type.
The following operations are only available for grid point clouds:
• Edit 2D: Edits the grid.
• Extract edges: Creates a polyline from the computed orientation and depth discontinuity during the pre-
processing phase. The polylines are as contiguous as possible. Adjust the max search distance and minimum
edge length to optimize the outcome.
• Cross Sections
• Multiresolution Mesh
• Simplify Points: Determine the most relevant points for the purpose of shape description, saving them into a
new unstructured point cloud.
Both types of point cloud have the following editable properties:

• Color ID: Color ID of the cloud. To render the items with their color IDs, select Navigation > Color by ID.
• Draw coordinate frame: Draws the local coordinate frame.
• Point size: Size of the drawn points.
• Subsampling: Subsampling rate for rendering of the cloud of points.
• Color mapped: Sets the color type to render the point cloud.

Advanced Options
A grid cloud of point can be viewed as “image” where the number of pixels (pix) are equal to the number of point the
scanner has acquired (valid and invalid).
Point clouds data storage

Data is stored in an XML description file + pure binary vector files. The file extension of the binary data specifies the
format, where data.NT: N=num of components (1,2,3,4…), T=data type (byte, short, word, half, float, double).
For point colors, a special 3bc (c stands for compressed dynamic range) file is created for rendering only, for instance
after histogram equalization.

For example, the unstructured point cloud myUnstruct will be saved in the following files in the project folder:
• Unstructs/myUnstruct.rup, an XML file containing the cloud's properties
• Unstructs/myUnstruct_rup/points.3f
• Unstructs/myUnstruct_rup/status.1i
• Unstructs/myUnstruct_rup/Reflectance.1f
• Unstructs/myUnstruct_rup/Reflectance.3bc
• …and more.
The grid point cloud myGrid will be saved in a similar way in the folder Grids inside the project folder. Grid point cloud
files have the extension RGP.

Polylines
Polylines are sets of open or closed polygons that appear in your 3D scene. You can create polylines in several ways:
• From the points you have collected in the point list window.
• From the cross sections tool (the polylines created by this tool are saved in your project in the Cross
sections group).
• From the context menu of a triangle mesh, using the command Tools > Meshing > Get mesh borders as
polylines.
• From the Mesh ridges & valleys tool.
• From the edge extractor of grid point clouds, available via Tools > Filtering > Extract edges.
The most important contextual command for polylines is the Export command, which allows you to export the polyline
in the DXF format, readable by AutoCAD®.
It is often useful to define polylines in order to delimit the region in which you want to calculate volume, or cut and fill
volume. Polylines can be also used to select points or to cut meshes.

Projectors
A projector is a camera view plus an image to project on the scene. It can colorize whatever geometry is illuminated by
the projector, which also allows you to occlusions if shadow mapping is supported.
The specific commands are:

• Load projector: Loads the image on the 3D model.
• Set projector image: Sets the desired image to project
• Edit projector image: Edits the projector’s image to set the transparent pixels in order to produce better blending
among the projectors
Enabling the Activate property loads the image and projects it on the scene
Shadow mapping
This technique works only for triangle meshes. To compute the occlusions from the viewpoint of the projector, a depth
buffer of the scene is created. The accuracy of the shadows depends on the sampling of this depth buffer, and therefore
on its width and height and depth range. Ideally the size should be infinite, but this cannot fit in the memory so the
resolution must be tuned. It's recommended that you use at least the width and height of the projected image and to
clip the near and far planes as close as possible to the desired scene's depth range. Because of this coarse resolution, the
resulting depth buffer is affected by quantization noise. To compensate for this, an epsilon term is used as a tolerance
for the depth comparisons. For each depth point, the epsilon formula (known as polygon offset) is ShadowScale∙Depth +
ShadowBias. If some random black points are visible when casting the shadows, try to increase the scale and bias, but
note that excessive values produce fewer shadows.
• Shadow bias: Additive term for polygon offset during the depth buffer generation (default is 16)
• Shadow scale: Multiplicative factor for polygon offset during the depth buffer generation (default is 2.5)
• Shadow width/height: Size of depth buffer used to compute the occlusions from the viewpoint of the projector. It's
recommended to use a size that is at least as big as the projected image, displayed in the tooltip of the projector.

Triangle Meshes
A triangle mesh is a 3D model represented by a set of triangles connected by common edges and common vertices. If a
point cloud includes a set of points, a mesh includes more information because it stores how the points are connected.
A triangle mesh therefore defines a surface in the 3D space. The mesh's edges that are owned by only one triangle are
called mesh borders. A mesh is called closed if there are no borders. A mesh is called manifold if all its edges are shared
by at most two triangles, and a mesh is called non-manifold if there are edges that belong to three or more triangles. A
mesh is called watertight if it has no holes, or if its borders are only one closed outer polygon.
To each vertex of the mesh, a color can be associated. Each mesh's triangle is rendered with a gradient of the three
colors of its vertices. The rendering color may depend also on the triangle's inclination with respect to the light source.
ATLAScan has various techniques for generating and editing meshes. Each triangle mesh item can contain more than one
submesh.
Actions available from meshes' context menu

• Meshing > Compute normals: Computes or updates the triangles' normals for the mesh.
• Meshing > Invert winding: Inverts the ordering of the vertexes for each triangle, so the surface is flipped to
the opposite side and the normals are inverted.
• Meshing > Compute area: Returns the mesh area as the sum of the areas of all the mesh's triangles.
• Meshing > Compute volume from Z=0 plane: Returns the volume resulting from integrating the mesh on the
XY plane of the current UCS. Mesh triangles below the XY plane will result in zero volume.
• Meshing > Get mesh borders as polyline: Creates a new polyline containing the mesh's borders and adds it to
the project.
• Meshing > Make single mesh: Clusters an arbitrary set of triangle meshes in a single mesh.

• Meshing > Convert to point cloud: Creates an unstructured point cloud from the vertices of the mesh using
the color attribute of the mesh.
• Meshing > Mesh editor: Opens the mesh editor dialog.
• Photo & Color > Sample texture at vertexes: Sample the color of the texture (if available) at the vertexes of
the mesh and stores it as the color attribute of the mesh.
• Photo & Color > Create texture map: Computes a blended texture map for the mesh from the projectors.
Meshes' editable properties
• Color ID: False color that you can activate by selecting Screen settings > Color by ID command.
• Draw coordinate frame: Draws the local coordinate frame.
• Point size: If the draw geometry is set to points, they are draw with this size.
• Draw texture: Loads and renders the texture (if one is available).
• Draw light: Renders the triangles (if their normals are valid) with smooth shading.
• Cull face: Renders the triangles only if the normal faces the view direction and hides backface triangles. If this
is set to False, both sides of the mesh are visible.
• Draw geometry: Displays the mesh as flat, wireframe (only edges), or only vertexes (points).

Advanced Options
Triangle mesh data storage

Data is stored in an XML description file + pure binary vector files subdivided in folders for each submesh. The file
extension of the binary data specifies the format, where data.NT: N=num of components (1,2,3,4…), T=data type (byte,
short, word, half, float, double). For example, if the mesh is called myMesh, it will typically be stored using the following
files in the project folder:
• Meshes/myMesh.rtm
• Meshes/myMesh_rtm/submesh/vertexes.3f
• Meshes/myMesh_rtm/submesh/triangles.3i
• Meshes/myMesh_rtm/submesh/...
See also Meshing techniques and Mesh editing.

User Coordinate System (UCS)

User coordinate systems are a central concept in ATLAScan. In the same project, many coordinate systems can exist. An
empty project contains at least the default UCS, called Main.
To create a new UCS, you can click on any project item and select Create UCS from this pose. The new UCS is added to
the User Coordinate Systems group in the project window. A UCS can be also created from the registration between two
point lists.
Current UCS
The Figures 1, 2 and 3 below show the same scene with three different UCS's set as the current one. The current UCS is
visible as the long coordinate axes (the red is the X axis, green is the Y, blue is the Z axis).
Figure 1 Figure 2 Figure3
There is always one and only one current UCS. This is the UCS displayed in bold in the User Coordinate Systems group in
the project window. To change the current UCS, right-click on another UCS and select Set as current.
The current UCS is the current coordinate system for all the operations in ATLAScan, such as the following:
• Point coordinates in the Readout Window are displayed in the current UCS.
• Model position in the Adjust Pose window is displayed in the current UCS.
• Points in the Point List Window are displayed, imported and exported in the current UCS.
• All project data (point clouds, meshes, polylines, etc.) are imported interpreting their position in the current UCS,
and exported saving their position as in the current UCS.
• The pose matrix shown in the Pose Dialog is referred to the current UCS.
• The navigation system will perform “human” rotation movements assuming that the horizontal plane is the XY plane

of the current UCS (see the option Enable human movements while rotating in Navigation Options for more
information).
• Many ATLAScan functions (such as Cut and fill calculation, for example) require a vertical direction and altitudes to
be defined in your 3D scene. ATLAScan generally assumes that the current UCS's Z axis defines the vertical direction,
and the altitudes are the distances of project items from the Z=0 plane of the current UCS.
• The Create/Edit Plane command makes a plane horizontal or vertical. The horizontal and vertical directions are,
again, always defined by the current UCS.
Therefore, if the user changes the current UCS, all the above-mentioned windows and functions will be affected in the
way they visualize the coordinates or compute the models' positions.

Create UCS from this pose

The context command Create UCS from this pose allows to create a new reference system (UCS) according to any item
in the project.
This is useful for creating a UCS according to an object in order to help with a particular visualization. For example you
could use a vertical plane fitting a wall to create a point of view (and thus an orthocamera) in front of the façade
following these steps:
• Define the plane
• In the plane's context menu, click Create UCS from this pose. A new UCS is created and stored in the User
Coordinate Systems folder.
• Select the new UCS and click Set as current. The new UCS now appears in bold and is used as the current reference
system.


ATLAScan Help

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

ATLAScan Help

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ATLAScan Help

Uploaded by

Copyright:

Available Formats

Optech ATLAScan Help

Preliminary draft for ATLAScan 1.0.0.81.

Preliminary draft for ATLAScan 1.0.0.81.

Recent camera positions.......................................................................................54

Preliminary draft for ATLAScan 1.0.0.81.

Manage units of measures ...................................................................................... 115

Preliminary draft for ATLAScan 1.0.0.81.

Select ............................................................................................................... 177

Preliminary draft for ATLAScan 1.0.0.81.

Windows ................................................................................................................... 240

Preliminary draft for ATLAScan 1.0.0.81.

Please click on one of the following sub-topics:

Preliminary draft for ATLAScan 1.0.0.81.

i. An Introduction on installing and licensing the software

Preliminary draft for ATLAScan 1.0.0.81.

System requirements — Minimum

System requirements — Recommended

Preliminary draft for ATLAScan 1.0.0.81.

1. Run the installation application.

3. Accept the License Agreement terms and click Next.

Preliminary draft for ATLAScan 1.0.0.81.

Preliminary draft for ATLAScan 1.0.0.81.

6. Choose a Start Menu Folder and click Next.

Preliminary draft for ATLAScan 1.0.0.81.

Setting up your environment

Preliminary draft for ATLAScan 1.0.0.81.

Activating the Software License

1. Open ATLAScan on the computer that you wish to license.

3. Click Create .c2v license file in the License Manager window.

Preliminary draft for ATLAScan 1.0.0.81.

ATLAScan is now licensed on the computer.

Updating the Software License

1. Open ATLAScan on the computer that you wish to license.

Preliminary draft for ATLAScan 1.0.0.81.

Copyright © 2017 Teledyne Optech Incorporated

Disclaimer and Limitation of Liability

See also the End User License Agreement.

Preliminary draft for ATLAScan 1.0.0.81.

Preliminary draft for ATLAScan 1.0.0.81.

Recent Projects Window

Recent Projects tools

Preliminary draft for ATLAScan 1.0.0.81.

Project Info Window

Project Info tools

Preliminary draft for ATLAScan 1.0.0.81.

Preliminary draft for ATLAScan 1.0.0.81.

Control Window, Simple

Preliminary draft for ATLAScan 1.0.0.81.

Control Window, Advanced

Preliminary draft for ATLAScan 1.0.0.81.

Preliminary draft for ATLAScan 1.0.0.81.

Preliminary draft for ATLAScan 1.0.0.81.

Create New Project Window

Preliminary draft for ATLAScan 1.0.0.81.

Preliminary draft for ATLAScan 1.0.0.81.

a. If you’re importing a point cloud the Process environment will be opened.

See also File menu.

Preliminary draft for ATLAScan 1.0.0.81.