Huawei SDC Technical White Paper - 290819

Huawei SDC
Technical White Paper
Issue V2.1
Date 2019-05-25
Huawei Technologies Co., Ltd.

Copyright © Huawei Technologies Co., Ltd. 2019. All rights reserved.
No part of this document may be reproduced or transmitted in any form or by any means without prior
written consent of Huawei Technologies Co., Ltd.
Trademarks and Permissions
and other Huawei trademarks are trademarks of Huawei Technologies Co., Ltd.
All other trademarks and trade names mentioned in this document are the property of their respective
holders.
Notice
The purchased products, services and features are stipulated by the contract made between Huawei and
the customer. All or part of the products, services and features described in this document may not be
within the purchase scope or the usage scope. Unless otherwise specified in the contract, all statements,
information, and recommendations in this document are provided "AS IS" without warranties, guarantees or
representations of any kind, either express or implied.
The information in this document is subject to change without notice. Every effort has been made in the
preparation of this document to ensure accuracy of the contents, but all statements, information, and
recommendations in this document do not constitute a warranty of any kind, express or implied.
Huawei Technologies Co., Ltd.

Address: Huawei Industrial Base
Bantian, Longgang
Shenzhen 518129
People's Republic of China
Website: http://www.huawei.com
Email: support@huawei.com
Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. i

Huawei SDC
Technical White Paper Contents
Contents
Change History ...............................................................................Error! Bookmark not defined.

1 Product Overview ......................................................................................................................... 1
1.1 Introduction .................................................................................................................................................................. 1
1.2 SDC Components ......................................................................................................................................................... 2
2 Related Technologies ................................................................................................................... 4

2.1 HD ................................................................................................................................................................................ 4
2.1.1 H.265 Codec Technology........................................................................................................................................... 4
2.1.2 SEC ............................................................................................................................................................................ 5
2.1.3 Corridor Mode ........................................................................................................................................................... 5
2.1.4 High-Frame-Rate Encoding Technology ................................................................................................................... 7
2.1.5 P-Iris .......................................................................................................................................................................... 8
2.1.6 Five Streams .............................................................................................................................................................. 8
2.2 Intelligence ................................................................................................................................................................... 9
2.2.1 Intelligent Encoding ................................................................................................................................................... 9
2.2.2 Abnormal Event Detection....................................................................................................................................... 11
2.2.3 Metadata .................................................................................................................................................................. 15
2.2.4 Adaptive Bandwidth ................................................................................................................................................ 16
2.2.5 Stream Smoothing.................................................................................................................................................... 17
2.2.6 iPCA ........................................................................................................................................................................ 18
2.2.7 SmartIR .................................................................................................................................................................... 20
2.2.8 ePTZ Technology ..................................................................................................................................................... 21
2.2.9 ROI .......................................................................................................................................................................... 22
2.2.10 Face Detection and Capture ................................................................................................................................... 23
2.2.11 Face Match ............................................................................................................................................................. 23
2.2.12 License Plate Recognition ..................................................................................................................................... 24
2.2.13 Fast Movement Detection ...................................................................................................................................... 24
2.2.14 Audio Detection Enhancement .............................................................................................................................. 24
2.2.15 Scene Change Detection ........................................................................................................................................ 24
2.2.16 Defocus Detection ................................................................................................................................................. 24
2.2.17 Visible Area ........................................................................................................................................................... 25
2.2.18 Video Quality Diagnosis ........................................................................................................................................ 26
2.2.19 Multi-Camera Collaboration .................................................................................................................................. 27
Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. ii

Huawei SDC
2.2.20 Software-Defined Camera ..................................................................................................................................... 28

2.2.21 Parking Detection .................................................................................................................................................. 29
2.2.22 Head Counting ....................................................................................................................................................... 29
2.2.23 Queue Length Detection ........................................................................................................................................ 29
2.2.24 Facial Attribute Recognition .................................................................................................................................. 29
2.2.25 Automatic Tracking ............................................................................................................................................... 30
2.2.26 Multi-Frame WDR ................................................................................................................................................. 30
2.2.27 Scene Adaptation ................................................................................................................................................... 30
2.2.28 ITS ......................................................................................................................................................................... 30
2.2.29 Vehicle Event Detection ......................................................................................................................................... 33
2.2.30 Smart Tracking....................................................................................................................................................... 33
2.2.31 1 + N (Master-Slave Camera Surveillance) ........................................................................................................... 35
2.2.32 Multiple-Algorithm Concurrency .......................................................................................................................... 36
2.2.33 Heat Map ............................................................................................................................................................... 36
2.2.34 MAC Address Collection ....................................................................................................................................... 37
2.2.35 T-Shot Intelligent Snapshot Engine ....................................................................................................................... 37
2.2.36 Traffic Data Collection .......................................................................................................................................... 37
2.2.37 Dual-Frame Integration.......................................................................................................................................... 38
2.3 Security ....................................................................................................................................................................... 38
2.3.1 Digital Watermark and Media Security Technologies .............................................................................................. 38
2.3.2 Video Buffering ....................................................................................................................................................... 40
2.3.3 Video Buffering Enhancement ................................................................................................................................. 41
2.3.4 IP Address and MAC Address Filtering ................................................................................................................... 41
2.3.5 IEEE 802.1X Access ................................................................................................................................................ 41
2.3.6 TCP Congestion Control .......................................................................................................................................... 42
2.3.7 KMC ........................................................................................................................................................................ 42
2.3.8 Security Certification ............................................................................................................................................... 43
2.3.9 GB 35114 ................................................................................................................................................................. 43
2.3.10 Intrusion Detection ................................................................................................................................................ 44
2.4 Reliability ................................................................................................................................................................... 44
2.4.1 Three Systems and Three Configurations ................................................................................................................ 44
2.4.2 Wide Temperature and Voltage Ranges ................................................................................................................... 44
2.4.3 Ingress Protection Rating ......................................................................................................................................... 44
2.4.4 IK10 Impact Protection Rating ................................................................................................................................ 46
2.4.5 Class D Anti-Corrosion ............................................................................................................................................ 46
2.4.6 Dual System Backup ................................................................................................................................................ 47
2.5 Ease of Use ................................................................................................................................................................. 47
2.5.1 ABF.......................................................................................................................................................................... 47
2.5.2 OEC ......................................................................................................................................................................... 47
2.5.3 GPON Access .......................................................................................................................................................... 49
2.5.4 Openness Design ..................................................................................................................................................... 49
2.5.5 Adaptive Rapid Deployment .................................................................................................................................... 50
Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. iii

Huawei SDC
2.5.6 Multiple Power Supply Modes ................................................................................................................................ 51

2.5.7 PoE, PoE+, and PoE++ ............................................................................................................................................ 51
2.5.8 Long-Distance Ethernet Transmission ..................................................................................................................... 52
2.5.9 Automatic Preventive Maintenance and Fault Reporting ........................................................................................ 52
2.5.10 Remote Zoom and Focus Control .......................................................................................................................... 53
2.5.11 G-sensor Image Stabilization ................................................................................................................................. 53
2.5.12 3D Positioning of High-Speed PTZ Dome Cameras ............................................................................................. 54
2.5.13 Stream Profiles ....................................................................................................................................................... 55
2.5.14 Active Anti-Fogging .............................................................................................................................................. 56
2.5.15 One-Click Log Collection ...................................................................................................................................... 56
2.5.16 Troubleshooting ..................................................................................................................................................... 56
2.5.17 Invisible IR Illumination ........................................................................................................................................ 57
2.5.18 GPS ........................................................................................................................................................................ 57
2.5.19 Posture Detection ................................................................................................................................................... 57
3 Appendix A Acronyms and Abbreviations............................................................................ 59
Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. iv

Huawei SDC
Technical White Paper 1 Product Overview
1 Product Overview
1.1 Introduction
Based on extensive and accumulated experience on video and audio technologies and video
surveillance system integration as well as technical features of network cameras in the
security industry, Huawei rolls out IPC6000 series, M series, and X series Software-Defined
Cameras (SDCs). These cameras are mission-critical products in the Huawei video
surveillance solution.
Huawei IPC6000 series, M series, and X series SDCs are video surveillance product series for
indoor and outdoor surveillance. They integrate a variety of functional modules such as video
compression, network transmission, intelligent algorithms (including license plate recognition,
face detection, facial recognition, facial attribute recognition, and personal attribute
recognition), and alarm handling. Those cameras adopt the embedded Linux operating system
and an embedded web communication module. They provide live video viewing and
pan-tilt-zoom (PTZ) camera control functions over networks (LAN, Internet, or wireless
network) to implement all-dimensional video surveillance. Huawei cameras also provide other
functions such as intelligent analysis, voice intercom, alarm input, relay output, motion
detection, analog video output, SFP interface, and local recording storage on SD cards. Those
cameras can work with the network video storage and recording system and management
platform software to construct a large-scale and distributed intelligent video surveillance (IVS)
system.
Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 1

Huawei SDC
1.2 SDC Components

Figure 1-1 SDC components
Image sensor (CCD or COMS)
Compression algorithm
Ethernet port
(RJ45, etc)
processing unit
CPU
Digital image
encoder
Video output port
(CVBS, etc)
Alarm port
Flash
DRAM Other ports (MIC
memory
and SD card, etc)
Lens ICR module External ports
As shown in Figure 1-1, an SDC is a highly integrated device that consists of the lens, IR-Cut
Filter Removable (ICR) module, image sensor, digital image encoder, compression algorithm
module, CPU, network unit, and local storage unit.
A scene (monitored object) is imaged on the image sensor through the lens and then
compressed to streams of a specified format (for example, H.264 or H.265) through the
encoder. The CPU then sends the streams and instructions to the surveillance center through
the network unit to implement video surveillance functions such as live video viewing,
recording storage, and recording playback.
Figure 1-2 Video surveillance services
Huawei SDCs adopt the integrated design and integrate a variety of functions such as
intelligent analysis (intelligence), real-time stream transmission (transmission), network
control (control), storage management, policy management (management), alarm
management, and SFP module. Additionally, the SDCs support multiple mainstream network
communication protocols such as ONVIF and TEYES.
Huawei SDCs use the embedded operating system and can work independently without the
assistance of computers. Therefore, the SDCs feature high integration and can significantly
enhance surveillance site deployment flexibility and system integration capabilities in new

Huawei SDC
and reconstructed video surveillance projects. The SDCs support remote maintenance and
alarm reporting, enhancing reliability of the entire video surveillance system and efficiently
reducing project implementation and O&M costs.

Huawei SDC
Technical White Paper 2 Related Technologies
2 Related Technologies
To adapt to various complicated and harsh environments, Huawei SDCs adopt a series of
technologies and means that comply with international standards, effectively enhancing the
camera reliability.
At the fabrication processing layer, with proper material selection and surface design, Huawei
cameras achieve the optimal heat dissipation, mould proof, salt spray proof, moisture proof,
and shock proof effect, reduce pollution to the environment, and enhance adaptability to
various common environments.
At the device hardware layer, Huawei provides surge protection, EMC protection system,
thermo design, and power supply design solutions and complies with international universal
standards to design SDCs. This helps enhance the camera reliability, reduce failure points, and
avoid complicated maintenance.
At the network transmission layer, the video buffering technology ensures video data integrity
in case of network outage.
At the data storage layer, the video buffering and digital watermark technologies ensure
surveillance data storage security.
At the overall system layer, the cameras adopt modular design, ensuring high product
performance and reliability.
2.1 HD
2.1.1 H.265 Codec Technology
High Efficiency Video Coding (HEVC), also known as H.265, is a video compression
standard, designed as a successor to the widely used AVC (H.264 or MPEG-4 Part 10). H.265
uses cutting-edge technologies to improve the relationship among the bit rate, encoding
quality, delay, and algorithm complexity to achieve the optimal settings. H.265 focuses on
increasing the compression ratio, enhancing robustness and fault rectification capabilities,
reducing the real-time delay, channel resource obtaining time, and random access delay, and
lowering the complexity. Thanks to algorithm optimization, H.264 can transfer SD digital
images at a bit rate lower than 1 Mbit/s. H.265 can transfer 720p (1280 x 720 pixels) common
HD audio and video at a bit rate ranging from 1 Mbit/s to 2 Mbit/s.
Compared with H.264, H.265 provides more tools to reduce the bit rate. In terms of the
encoding unit, each macroblock (MB) in H.264 is fixed at 16 x 16 pixels, while each

Huawei SDC
macroblock in H.265 ranges from 8 x 8 pixels to 64 x 64 pixels. For areas with unobvious
color variation (for example, red vehicle body and gray ground), the macroblocks segmented
are relatively larger and there are fewer codes after encoding. For areas with more details (for
example, tyre), the macroblocks segmented are relatively smaller and there are more codes
after encoding. In this case, key areas of images are encoded, reducing the overall bit rate and
enhancing encoding efficiency. Additionally, the intra-frame prediction mode of H.265
supports 33 types of directions (H.264 supports only eight types) and provides better motion
compensation and vector prediction methods.
H.265 is aimed to transfer higher quality network video at a limited bandwidth. Compared
with H.264, H.265 reduces half of required bandwidth resources while producing the same
image quality. In conclusion, H.265 enhances codec efficiency, saves transmission bandwidth
and storage space, and provides technical basis for future higher video resolution
development.
2.1.2 SEC
During video data transmission over networks, if a network exception, unstable
communication, or high packet loss rate occurs, a series of problems may occur on the site,
for example, the received data packet is incomplete, video cannot be properly decoded or
played back, or artifacts or frame freezing occurs on video images.
Huawei SDCs support patented Super Error Correction (SEC) technology. With the
technology, a camera sends data with error correction codes. The recipient performs error
detection on the received data based on the error correction codes. If an error is detected, the
recipient uses the SEC recovery algorithm to recover lost packets, enhancing video image
effect and preventing image artifacts and frame freezing. With the SEC technology, Huawei
SDCs can ensure proper image display (without artifacts or frame freezing) in the case of up
to 20% packet loss rate.
Figure 2-1 SEC technology implementation diagram
Orig inal data stream

Packet loss
Networ k Packet
loss Recove ry
SEC encoding
The ca mer a colle cts and send s data. The clie nt receives the data wi th packet loss.
Note: 1-1->1-2: no SEC processi ng 2-1->2-2: SEC processi ng added
To use the SEC function, cameras must support media transmission with the platform in SEC mode.
2.1.3 Corridor Mode

The aspect ratio of common cameras is usually 16:9, 5:4, or 4:3. Figure 2-2 shows the video
image in the 16:9 aspect ratio. As shown in Figure 2-2, most part of the image shows the walls
which are useless for video surveillance. What's worth is the corridor passage, which occupies
only about 25% of the effective image pixels, causing resource wastes for video image
compression, transmission, and storage.

Huawei SDC
Figure 2-2 Video image in 16:9 aspect ratio
Huawei SDCs support the 9:16 aspect ratio. In narrow, high, and vertically long scenes, the
corridor mode allows a camera to rotate the video image by 90 degrees, which reduces the
portion of the walls on the video image and focuses on the corridor passage. This ensures that
the effective surveillance area accounts for about 50% of the total image pixels.
Figure 2-3 Video image in 9:16 aspect ratio

Huawei SDC
2.1.4 High-Frame-Rate Encoding Technology

Any video file is composed of consecutive images. An image is called a frame. For example, a
video feed of 1s long is composed of 25 consecutive images. In this case, the frame rate is 25
fps. In PAL mode, a 25 fps video feed can present motions vividly. If the frame rate is less
than 25 fps, the motions in the video will not be smooth. A lower frame rate indicates poorer
image smoothness (even jumping movements may occur). Therefore, the frame rate plays an
essential role in video surveillance.
The frame rate requirements on HD cameras vary in different industries. For example, in the
transportation industry, there are strict requirements on the frame rate in road checkpoint
surveillance, lane surveillance, and even casino surveillance. Cameras are the most important
devices in the video surveillance system. The special market requirements of cameras will
definitely boost camera technology updates and further promote in-depth video surveillance
application development.
With the high-frame-rate encoding technology, Huawei SDCs can implement HD encoding at
60 fps in addition to common HD encoding at 25 fps or 30 fps, delivering smoother video
images and sharply enhancing HD video experience.
Figure 2-4 HD video at 30 fps

Huawei SDC
Figure 2-5 HD video at 60 fps
As shown in Figure 2-4 and Figure 2-5, in the same scene, the video image at 60 fps is smoother than
that at 30 fps.
2.1.5 P-Iris
The motor in the P-Iris lens precisely controls the position of the iris opening. Together with
the algorithm for optimizing the performance of the lens and image sensor, the P-Iris
automatically provides the best iris position for optimal image quality in all lighting
conditions. In bright situations, the P-Iris limits the closing of the iris to avoid blurring
(diffraction) caused when the iris opening is too small, delivering images with better contrast,
clarity, resolution, and depth of field (DoF).
 This feature applies to IPC6681-Z20, IPC6285-VRZ, IPC6285-VMZ, and M series
bullet cameras.
 The P-Iris function is enabled by default on the camera web page. Users can set P-Iris
parameters to enable manual iris control.
2.1.6 Five Streams

Generally, a common camera can provide triple streams, while a Huawei camera can provide
a maximum of five streams (one primary stream + four secondary streams) at the same time.
Users can log in to the platform or camera web system to set video stream parameters. As
shown in Figure 2-6, users can set video stream parameters for each stream separately.

Huawei SDC
Figure 2-6 Setting video stream parameters
This feature applies to M12 and M22 series cameras.

Maximum five-stream capability supported by M12 and M22 series cameras:
1080p@30 fps + 1080p@30 fps + 720p@30 fps + D1@30 fps + D1@30 fps
2.2 Intelligence
2.2.1 Intelligent Encoding
The intelligent encoding of most manufacturers in the industry is implemented through
dynamic GOP (reducing the bit rate) and dynamic ROI (improving the image quality of
moving objects). The I-frame in the dynamic GOP will cause large fluctuation of the overall
bit rate and pose higher network and decoding requirements. As for dynamic ROI, when an
object moves fast, the ROI cannot catch up with the object, and obvious breathing effects can
be detected on the edge of the ROI.
Huawei implements intelligent encoding by adjusting the internal encoding policy in the
encoder, which helps to decrease the bit rate and improve the image quality. The applied
technologies are as follows:
1. Adaptive variable bit rate (VBR) control algorithm
2. Intelligent reference frame mode and virtual I-frame
Adaptive VBR control

Huawei SDC
Adaptive VBR control allows bit rate fluctuation during bit rate collection to ensure stable
quality of encoded images.
The bit rate control algorithm detects the object status (moving or static) in the current scene,
uses a higher bite rate for encoding when an object is moving, and decreases the bit rate when
the object becomes static. Bit rate control is performed inside the encoder. The algorithm
determines the scene based on the motion amount detected during encoding and then adjusts
the bit rate control policy.
Compared with common VBR control, adaptive VBR control, while ensuring the image
quality, can effectively decrease the bit rate when an object is static and gradually increase the
bit rate when an object is moving. Adaptive VBR control features higher real-time
performance, progressive bit rate control, and excellent image quality.
Intelligent reference frame mode and virtual I-frame
In intelligent reference frame mode, the P-frame references the Instantaneous Decoder
Refresh (IDR) frame (long-term reference frame) and forward reference frame (short-term
reference frame, that is, virtual I-frame). The temporal correlation between the two reference
frames is used to improve the encoding and compression performance. The intelligent
reference frame mode is mainly applied in video surveillance scenarios where cameras are
installed in fixed positions and there are both static and moving persons and objects.
For a static ROI, the temporal correlation between the long-term reference frame and the
current frame can be used to decrease the bit rate significantly and reduce the breathing and
trailing effects.
For a moving ROI, the short-term reference frame is used to perform motion estimation. In
intelligent reference frame mode, the IDR-frame interval is prolonged and virtual I-frames are
inserted periodically. This greatly decreases the bit rate in surveillance scenarios and improves
the image quality.

Huawei SDC
2.2.2 Abnormal Event Detection

The rapid video surveillance deployment in recent years has seen a growing number of
surveillance sites. A large amount of video data causes severe challenges to live video
surveillance and video data usage. A large video surveillance system usually connects to
hundreds of to tens of thousands of cameras. It is difficult for surveillance operators to view
each channel of video to detect abnormal events. The intelligent video analysis function can
help solve the issue. With the intelligent video analysis function, SDCs can extract and record
events that users are interested in to achieve real-time alarm reporting or efficient post-event
analysis, enhancing security efficiency.
Huawei SDCs support event analysis functions (such as tripwire crossing detection, loitering
detection, intrusion detection, abandoned object detection, removed object detection, and
object classification) and can automatically detect key information in video, reducing labor
costs, enhancing surveillance efficiency, and meeting intelligent analysis requirements in
different scenarios.
Huawei SDCs can work properly in all-weather conditions such as environments with natural
light and low-light environments with infrared light radiation. The deep learning technology is
embedded into cameras, avoiding false positives caused by dynamic backgrounds, such as
waving tree branches and water rippling as well as interference of lightning, shifting between
light and shade, specular reflection, harsh weather conditions, and intermittent movement.
The camera can cover a detection distance of up to 150 meters and monitor a maximum of 30
objects (people and vehicles) at the same time.
1. Tripwire crossing detection
Users can set a tripwire, either a straight line or polygon curve (multiple straight lines
combined), on the video image. The system can identify and track moving objects on
video images. When an object crosses the tripwire in a specified direction, the system
generates an alarm, and frames and tracks the object. Users can also flexibly set the
detection sensitivity.
Figure 2-7 Tripwire crossing detection
2. Loitering detection
Users can specify a surveillance area (rectangle or polygon) on the video image. When
an object remains in the surveillance area for a period longer than the preset duration, the
system generates an alarm, and frames and tracks the object. Users can also flexibly set
the detection sensitivity.

Huawei SDC
Figure 2-8 Loitering detection
3. Intrusion detection
an object enters the surveillance area, the system generates an alarm, and frames and
tracks the object. Users can also flexibly set the detection sensitivity.
Figure 2-9 Intrusion detection
4. Abandoned object detection

an object is abandoned in the surveillance area for a period longer than the preset
duration, the system generates an alarm, and frames and tracks the object.

Huawei SDC
Figure 2-10 Abandoned object detection
5. Removed object detection

an object is removed from the surveillance area for a period longer than the preset
duration, the system generates an alarm, and frames and tracks the object.
Figure 2-11 Removed object detection
6. Area entry detection

an object enters the surveillance area, the system generates an alarm and frames the
object. This function is used to protect key sites against intrusions. Users can also
flexibly set the detection sensitivity.

Huawei SDC
Figure 2-12 Area entry detection
7. Area exit detection

an object leaves the surveillance area, the system generates an alarm and frames the
object. This function is used to prevent valuable objects from being stolen, for example,
public goods, valuable assets, and vehicles. Users can also flexibly set the detection
sensitivity.
Figure 2-13 Area exit detection
8. Fast movement detection

an object moves fast in the surveillance area, the system generates an alarm and frames
the object. Users can also flexibly set the detection sensitivity.

Huawei SDC
Figure 2-14 Fast movement detection
9. Crowd gathering detection

Users can specify a surveillance area (rectangle or polygon) on the video image. The
system automatically counts the number of people in the surveillance area. When the
crowd density in this area exceeds the specified threshold, the system generates an alarm.
Figure 2-15 Crowd gathering detection
2.2.3 Metadata
To enable the platform to flexibly use the intelligent analysis functions of cameras, Huawei
SDCs extract various objects generated in intelligent analysis and package them into
composite streams. All the object information is called metadata.

Huawei SDC
The intelligent analysis metadata includes the object location (coordinates, width, and height),
type (people, vehicles, or articles), speed, color, contour, and background (such as width and
height).
Figure 2-16 Object classification
2.2.4 Adaptive Bandwidth

If the network conditions are poor, video images may not be smooth or frame freezing occurs
due to data congestion during network transmission. With the adaptive bandwidth technology,
cameras can lower the image clarity by automatically reducing the bit rate due to insufficient
network bandwidth, ensuring image smoothness and continuity.
The cameras check whether the network bandwidth is sufficient based on the packet loss rate.
If the network bandwidth is insufficient, the cameras dynamically set a bit rate that is lower
than the user-defined value, enabling the bit rate to adapt to the current network bandwidth. If
the network bandwidth is sufficient, the cameras dynamically increase the bit rate. The bit rate
upper limit is specified by users.

Huawei SDC
Figure 2-17 Adaptive bandwidth
Users set the bit rate.
Sufficient
bandwidth
Network
Insufficient
bandwidth
Lower the bit rate to adapt to the current

network bandwidth Transmission network
Artifacts and packet loss occur if the If adaptive bandwidth is enabled, the bit rate is
bandwidth is 3 Mbit/s for transmitting 1080p automatically adjusted to about 3 Mbit/s, solving
@ 4 Mbit/s video by default. artifacts and packet loss issues.
2.2.5 Stream Smoothing

On a network with fixed but insufficient bandwidth, great disparities between I-frames and
P-frames may result in sharp increase in the network traffic, causing live video freezing.
During I-frame transmission, some I-frames are incompletely transmitted due to fixed
bandwidth, resulting in artifacts. Huawei restricts the data transmission rate and suppresses
the peak bandwidth of media streams, reducing requirements on peak bandwidth from burst
bit rate of media streams and ensuring stable network bandwidth usage. A greater stream
smoothness coefficient indicates stricter data transmission rate control and more distinct
smoothness effect.
Figure 2-18 Stream smoothing
S tream sm oothing disabled Trans mis s ion

I-fram e P-fram e network
E ncoding
modul e
Transmission
network
Camera encoding
S tream sm oothing enabled
In the case of fixed network bandwidth, the system can perform traffic shaping on video data
to prevent burst network traffic during video transmission, ensuring video transmission
stability and reliability.

Huawei SDC
In actual surveillance scenarios, if an object in the surveillance view encounters big changes
or the surveillance view varies greatly due to PTZ device rotation, the data packet size of
I-frames after H.264 encoding will increase sharply and exceeds the value allowed by the
maximum bandwidth on the actual transmission network. The sharp video stream increase
will lead to network congestion, increasing the packet loss and affecting the surveillance
effect.
With the stream smoothing technology, Huawei SDCs can evenly send peak streams at frame
intervals based on user settings to prevent data packet loss due to burst streams, reducing
video streams' requirements on peak network bandwidth and ensuring video stream
transmission stability and reliability (as shown in Figure 2-19).
Figure 2-19 Stream smoothing bandwidth diagram
Maximum instantaneous value: about 900,000
1M suppression.pcap
Maximum instantaneous value: about 100,000
2.2.6 iPCA
Packet Conservation Algorithm for Internet (iPCA) is the first multiple-input-multiple-output
quality measurement technology in the industry, which solves the N2 connection issue in
traditional point-to-point quality measurement technologies (BFD, NQA, and Y.1731). iPCA
technology uses the enhanced area-based packet conservation mechanism to monitor the
quality of a connectionless network and also provides accurate fault locating capabilities.
Packet conservation indicates that the number of packets leaving a system (network, link,
device, or board) equals the number of packets arriving at the system. If data flows passing
through a system comply with packet conservation, packet loss does not occur and the packet
transmission quality is ensured. Currently, iPCA mainly measures the packet loss rate, which
is the most important factor that affects service experience.

Huawei SDC
Figure 2-20 iPCA measurement principle
Packets arriving at the system Packets leaving the system
Internally generated
packets
Measured system
(network/linke/device/board)
Pckets internally
terminated by the system
The iPCA quality measurement mechanism is simple. A measured system is in normal state if
the following condition is met: Number of packets arriving at the system + Number of
internally generated packets = Number of packets leaving the system + Number of packets
internally terminated by the system. If this condition is not met, some packets have been
dropped in the system.
The measurement of a device or link is iPCA device-level measurement. The measurement of
a network consisting of multiple devices is iPCA network-level measurement.
Figure 2-21 iPCA measurement solution
Solution Overview
Measurement Object:
iPCA monitors packet loss at the
device, link, or region level to
measure network quality.
Measurement Scenario:
Branch 2
Branch 1
Device-level monitoring:
Single agile device (excluding non
ENP and packet loss impacts)
Link-level monitoring
Direct link monitoring between
agile devices
HQ
Network-level monitoring
Devices surround the consecutive
domains consisting of non-agile
devices (including third-party devices)
End-to-end link that transmits specified
service flows

Huawei SDC
The iPCA algorithm is embedded in Huawei SDCs. The data packets can be monitored from
the beginning. If traffic packages are lost over the entire transmission link, frame freezing or
artifacts will occur. eSight can quickly locate failure points to help resolve issues.
[Application Condition]:
iPCA must be used with agile network switches and eSight.
2.2.7 SmartIR
A problem exists on the proportion between the high beam and the low beam for common
infrared cameras. If a camera works in telephoto (long-focus) mode, the radiation distance of
the low beam and the brightness of the high beam are insufficient. Scenes that users focus on
cannot obtain proper infrared radiation brightness. If a camera works in wide-angle (short
focus) mode, the brightness of the low beam is insufficient, while the high beam will result in
over-brightness in the central area of the image. This is called flashlight effect, as shown in
Figure 2-22.
Figure 2-22 Flashlight effect
With the SmartIR technology, Huawei SDCs can enable automatic exposure and set an
optimal proportion between the high beam and the low beam based on the current focal length
(zoom), image brightness, and gain. This helps ensure proper image brightness and even
infrared radiation for cameras that work in the range from wide-angle mode to telephoto mode.
and show the image effects in different infrared radiation distances.

Huawei SDC
Figure 2-23 Image effect in an infrared range of 30 m
Figure 2-24 Image effect in an infrared range of 70 m
2.2.8 ePTZ Technology

With the electronic pan-tilt-zoom (ePTZ) technology, Huawei SDCs allow users to perform
ePTZ operations (pan and zoom-in) on any area of interests, crop the area, zoom in on the
video image in the area, and send video streams of the selected area to the platform in
secondary stream mode. When users view surveillance recordings of a camera, they can
choose to view only secondary streams of the camera, which are transmitted to the platform
over networks, reducing the network bandwidth consumption and lowering requirements on
the storage capacity. Alternatively, users can enable two channels of video images, with one
channel for delivering panoramic video images and the other for delivering video images in a
key area.

Huawei SDC
Figure 2-25 ePTZ control
2.2.9 ROI
On the one hand, HD cameras increase the image clarity; on the other hand, they cause a
series of challenges to the network bandwidth and storage capacity for HD video surveillance
systems. To alleviate video data transmission and storage pressure and further promote HD
video surveillance system application in various sectors, Region of Interest (ROI) technology
is developed. Users are usually interested in a specified area on video images. The clarity of
the image in the key area can be higher than that in other areas. With the ROI technology,
Huawei SDCs can ensure effective surveillance over a specified area in the case of
insufficient network bandwidth.
Users can specify one or more areas on surveillance images as ROIs. The image quality in the
ROIs can be different (higher or lower) from that in other areas. That is, the system performs
near-lossless compression (high bit rate) on ROIs and lossy compression on non ROIs (low
bit rate), ensuring higher quality for reconstructed images and achieving a higher compression
ratio.

Huawei SDC
Figure 2-26 Huawei ROI configuration page
2.2.10 Face Detection and Capture

The camera can automatically detect faces that appear in the surveillance view, capture face
images, and upload the face images as metadata.
In the camera web system, users can view detected faces framed in rectangles in real time.
Figure 2-27 Face detection and capture
When a face keeps appearing, the camera will capture a face image with the best quality as
required and upload it as metadata. An image is considered of good quality if it is clear and
records the complete frontal face (with both eyes). Such an image is generally used by the
platform for face match or facial recognition.
2.2.11 Face Match

The camera can automatically detect faces that appear in the surveillance view, match
detected faces with blacklisted faces, and generate an alarm when the face similarity exceeds
the specified threshold.

Huawei SDC
In the camera web system, users can view detected faces framed in rectangles in real time and
the face match results.
Figure 2-28 Face match
2.2.12 License Plate Recognition

After a vehicle enters the view of a camera, the license plate recognition algorithm detects and
tracks the vehicle, and locates the license plate in the tracking process. After the license plate
is located, the camera performs preprocessing (for example, deblurring and slant correction)
on the license plate and then performs license plate recognition. These steps complement each
other, and the license plate recognition recall rate and accuracy in each step are recorded.
2.2.13 Fast Movement Detection

Users can specify a surveillance area (rectangle or polygon) on the video image. When the
pixel change of an object in the area exceeds the preset range, the system generates an alarm,
and frames and tracks the object.
2.2.14 Audio Detection Enhancement

In the camera web system, users can enable the audio existence detection function or sudden
volume increase or decrease detection function, and set the detection sensitivity parameters.
The audio existence detection function checks whether there is audio data input in real time.
The sudden volume increase or decrease detection function collects statistics on audio data in
the same interval. If the audio data difference (increase or decrease) between adjacent
intervals exceeds a specified threshold, an alarm is generated.
2.2.15 Scene Change Detection

The scene change detection function performs background modeling for the video image,
extracts the background image of the video image, compares images in different time
intervals using the corresponding image features, and finally determines whether the images
in different time intervals are the same. If they are different, the scene has changed.
2.2.16 Defocus Detection

The defocus detection function can be used to detect whether a video image is blurred due to
lens defocus and generate an alarm based on the detection result.
Currently, this function cannot determine whether an image is clear absolutely. When this
function is enabled, users need to calibrate the detection result.

Huawei SDC
1. If a user disables the defocus detection function and enables it again, the image produced
during the switching period is regarded as clear by default, based on which subsequent
detection is performed.
2. If the automatic focusing function is available, the image delivered after automatic
focusing is complete is regarded as clear by default, based on which subsequent
detection is performed.
3. If a user drags the focusing slider and the image after the slider dragging is clearer than
that after automatic focusing is complete, the image after the slider dragging is used as
the basis for subsequent detection.
2.2.17 Visible Area

Users want to know the surveillance coverage of a camera in an area and obtain the
surveillance blind spots. The visible area function can achieve the following objectives:
1. Automatically generates coordinates of surveillance sites.
2. Makes valid surveillance areas visible.
3. Checks blind spots in surveillance areas.
A camera uses SDK interfaces to provide the integrated platform with the camera focal length,
optical zoom ratio, field of view, angle between the lens orientation and the ground, sensor
size, and geographic location of the camera. The integrated platform uses the data to display
the surveillance area map and surveillance direction map of the camera and receives camera
location alarm messages. The camera provides only the awareness capability and needs to
work with the platform to implement the feature.
The electronic compass is used to locate the northbound direction. The horizontal and vertical
motors rotate to mark the azimuth and improve the azimuth precision in each direction.
Operation Guide
1. Choose Settings > PTZ > Location > PTZ Details. Set Display mode to Azimuth.
2. Choose Settings > System > Visible Area > Position. Use the electronic compass to
perform calibration. After the calibration is complete, you can obtain the actual azimuth.
Parameter Description
Installation Info Used to query the camera installation height.
Sensor Size Used to query the camera sensor size, including the horizontal and
vertical sizes.
Position Used to set the camera longitude, latitude, azimuth, and tilt angle.
Lens Info Used to query the camera lens information, including the zoom ratio,
digital zoom, focal length, horizontal field of view, and vertical field
of view.
Only IPC6681-Z20 can automatically obtain visible areas. Users need to manually mark visible areas for
other camera models.

Huawei SDC
2.2.18 Video Quality Diagnosis

The rapid video surveillance deployment in recent years has seen a growing number of
surveillance sites. A large video surveillance system usually connects to hundreds of cameras,
even tens of thousands. How cameras can automatically report faults and notify users of
camera detection and maintenance in a timely manner has become a major problem.
With the video quality diagnosis function, cameras can effectively detect a variety of image
abnormalities caused by various reasons such as lens blocking and man-made reasons. After
detecting an image abnormality, cameras can generate an alarm. This helps maintenance
personnel quickly locate faults and reduce losses caused by image abnormalities.
The video quality diagnosis system is an intelligent video fault analysis and warning system
that integrates multiple technologies such as image processing, computer vision, computer
graphics, and image analysis technologies. The system can analyze collected images and
extract valuable object features from the images. Through predefined video quality diagnosis
conditions and rules, the system can automatically generate alarms upon detecting image
abnormalities, such as image noise, stripe interference, image blur, color cast, frame freezing,
gain imbalance, and video signal loss.
Figure 2-29 shows the image abnormalities that Huawei SDCs can detect.
Figure 2-29 Video quality diagnosis

Huawei SDC
2.2.19 Multi-Camera Collaboration

If only one camera is deployed at a surveillance site (for example, an entrance) for facial
recognition, the recall rate is not high and some false positives may be reported. Multi-camera
collaboration means that multiple cameras are deployed in the same area to implement facial
recognition. A person who moves around the surveillance area will be identified by multiple
cameras. In this way, the recall rate of facial recognition is improved, and false positives are
reduced. The following figure shows the deployment and implementation of multi-camera
collaboration.
Figure 2-30 Multi-camera collaboration
The process is as follows:

1. The administrator imports the face blacklist on the web client of the camera deployed at
an entrance and creates an alert task.
2. The administrator enables multi-camera collaboration on the IVS platform, specifies the
cameras in a collaboration set and the entrance camera, and sets collaboration parameters
such as the recall rate alarm threshold and recall rate warning threshold. The IVS
platform delivers the parameters to the entrance camera.
3. When the entrance camera detects a face, the camera matches the face with the blacklist.
If the recall rate is higher than the alarm threshold, the system displays the result and
ends the recognition. If the recall rate is lower than the warning threshold, the system
ignores the recognition result. If the recall rate is lower than the alarm threshold but
higher than the warning threshold, the system initiates multi-camera collaboration and
reports collaboration data to the IVS platform. The reported data includes the facial
feature data, recall rate, and the matched blacklist records in this recognition, as well as
the alarm threshold and warning threshold of the recall rate.
4. After receiving the collaboration data, the IVS platform broadcasts the data to other
cameras in the collaboration set. After receiving the collaboration data, the collaborative
cameras set up an alert task based on the blacklist record matched by the entrance
camera.
5. When a collaborative camera detects a face, the camera matches it with the blacklist in
the collaboration data and obtains a recall rate. The final recall rate is a weighted value
of the recall rates obtained by the collaborative cameras and the entrance camera. If the
weighted recall rate is higher than the alarm threshold, the system outputs the result and
ends the current collaboration task. Or the system waits for the recognition result of other
collaborative cameras.

Huawei SDC
Multi-camera collaboration improves the overall recall rate and effectively reduces the false
positives.
2.2.20 Software-Defined Camera

Traditionally, users need to purchase different models of cameras for different purposes,
which results in overlapping device investment. Huawei's SDCs allow users to define cameras
based on the application scenarios to meet various service requirements, reducing device
costs.
The Device Management (DM) platform or camera web client can deliver intelligent
algorithms and models so that common cameras can be quickly switched to dedicated cameras,
adapting to different application scenarios and reducing costs.
Figure 2-31 Service flow of the SDC
Online Intelligent Algorithm Loading Process

1. The administrator uploads an intelligent algorithm (such as facial recognition, traffic
violation detection, and vehicle recognition algorithms) on the DM management page
and associates it with cameras.
2. The DM platform pushes the intelligent algorithm to the cameras. The cameras disable
the algorithm in use, write the new algorithm and model into the corresponding flash
partition, and then enable the new intelligent application.
3. The cameras report the loading and operating status of the intelligent algorithm to the
DM platform, and the DM platform then displays the information to the administrator.
4. The administrator can also upload the intelligent algorithm through the camera web
client. In this case, the algorithm will be delivered to and applied on the current camera
only.
Online Intelligent Algorithm Upgrade Process
1. The administrator uploads the new version of an intelligent algorithm on the DM
management page and associates the new version with cameras.

Huawei SDC
2. The DM pushes the new version of the intelligent algorithm to cameras. After receiving
the new algorithm version, the cameras disable the corresponding intelligent application
and start to update the algorithm. When the update is complete, the cameras restart the
corresponding intelligent application.
3. The administrator can also upload the new version of the intelligent algorithm through
the camera web client. In this case, the new version will be delivered to and applied on
the current camera only.
Intelligent Algorithm License Control Process
1. The administrator applies for licenses in batches using the ESN list on the license
platform.
2. The administrator imports the license files into the DM platform. The DM platform
parses the license files and associates them with the corresponding cameras based on the
ESNs.
3. The DM platform delivers the license files to the corresponding cameras.
4. The cameras save the license files to the local computer and start the licensed intelligent
application.
5. Each time a camera restarts, it reads the license file from the local computer and starts
the licensed intelligent application accordingly.
2.2.21 Parking Detection

The parking detection function can be used in traffic hubs such as departure halls at airports,
railway stations, bus stations, and passenger terminals, as well as areas where parking
violations frequently occur, such as gates of schools, shopping malls, restaurants,
supermarkets, sport venues, and streets. If a vehicle is parked in a surveillance area for a
period longer than the preset duration, the camera can automatically detect the vehicle and
report an alarm to related agencies and personnel.
2.2.22 Head Counting

The head counting function is mainly applied to entrances and exits of public places such as
shopping malls, supermarkets, chain stores, exhibitions, and museums. The camera must be
vertically installed, with a tilt angle of about 5° to 15°. Users can adjust the settings for head
counting, such as the line-crossing direction, detection sensitivity, and alarm threshold. If the
head count exceeds the threshold, an alarm will be generated. The head counting statistics can
be displayed in reports.
2.2.23 Queue Length Detection

The queue length detection function is primarily applied to entrances and exits of railway
stations, subway stations, shopping malls, and airports. The camera calculates the total queue
waiting duration based on the number of queuers detected and the predefined average service
duration for one person. When the total queue waiting duration exceeds the configured alarm
threshold, an alarm is generated.
2.2.24 Facial Attribute Recognition

Huawei SDCs designed for face detection or recognition provide the facial attribute
recognition function. Attributes of a face meeting certain pixel requirements can be identified.
The facial attributes include age, ethnicity, gender, mask wearing, and glasses wearing. The
facial attributes can be uploaded as metadata to the platform (through SDK interconnection)
or viewed on the web page.

Huawei SDC
2.2.25 Automatic Tracking

The automatic tracking function is mainly used for perimeter surveillance of airports, harbors,
campuses, power plants, scenic spots, or other areas. The automatic tracking modes include
the following:
1. Manually-triggered tracking: Users can specify an object in the live video image for
tracking. The camera will track the selected object at a configurable movement speed
and zoom ratio for a specified period.
2. Alarm-triggered tracking: Automatic tracking can be triggered by behavior analysis
alarms such as intrusion, loitering, and tripwire crossing.
3. Full-image automatic tracking: When the automatic tracking function is enabled, the
camera starts automatic tracking as long as a moving object appears in the surveillance
image. If there are multiple moving objects, the camera tracks the nearest or the fastest
object preferentially.
2.2.26 Multi-Frame WDR

Multi-frame WDR helps produce sharp images while increasing the brightness of dark areas
based on the high signal-to-noise ratio (SNR). To produce a WDR image, three frames at
different exposure levels are combined. This displays more details in a final image than the
dual-frame WDR mode, achieving a more natural transition between bright and dark areas.
2.2.27 Scene Adaptation

The cameras can determine the weather conditions, lighting conditions, and object movement
state in the surveillance scene in real time and adjust the imaging settings accordingly to
ensure the image clarity in all-weather conditions.
2.2.28 ITS
Intelligent Transportation Surveillance (ITS) cameras are used to detect various traffic
violations in urban roads or highways in real time. Common violations include red-light
running, end-number policy violation, speeding/low speed, not following lane markings,
wrong-way driving, unsafe lane change, parking in yellow zones, bus lane violation, motor
vehicles driving on non-motor vehicle lanes, and large vehicles driving on prohibited lanes.
ITS cameras can also collect traffic flow statistics, perform secondary vehicle feature
recognition, and take snapshots of in-vehicle violations (such as seat belt infractions and
hands-free device infractions).
1. Red-light running of motor vehicles
ITS cameras can detect red-light running in vehicle rear detection mode. This function is
primarily applied in urban intersections or other areas where traffic lights are deployed.
The implementation process is as follows:
1) When an ITS camera detects an object entering the surveillance area before the stop
line, the camera starts tracking the object immediately. If a red light is detected, the
camera takes a snapshot before the vehicle arrives at the stop line. If the camera
determines that a red-light running violation occurs, this snapshot will be the first
evidence image.
2) If the camera detects that the vehicle leaves the stop line at a red light, it takes the
second snapshot.
3) If the camera detects that the vehicle leaves the straight-through trigger line at a red
light, it takes the third snapshot.

Huawei SDC
In this way, a complete set of red-light running snapshots are taken. The camera will also
take a close-up snapshot, synthesize these snapshots into a violation image, and send the
image to the platform.
2. End-number policy violation
ITS cameras can detect vehicles on road that violate the end number policy.
When a vehicle passes, the camera recognizes the license plate of the vehicle and checks
the end number policy. If the vehicle is banned on roads according to the end number
policy, the camera takes snapshots, synthesizes the snapshots into a violation image, and
sends the image to the platform.
3. Speeding/low speed detection
ITS cameras can detect vehicles at a speed greater or less than is reasonable and prudent.
When a vehicle passes, the camera detects the vehicle speed. If the vehicle drives at a
speed greater or less than is reasonable and prudent, the camera takes snapshots,
synthesizes the snapshots into a violation image, and sends the image to the platform.
4. Motor vehicles not following lane markings
ITS cameras deployed in ePolice mode can detect vehicles not following lane markings.
When a vehicle passes, the camera takes a snapshot of the vehicle. When detecting that
the movement path of the vehicle is against the lane markings, the camera records a
violation, takes snapshots, synthesizes the snapshots into a violation image, and sends
the image to the platform. Not following lane markings includes not following the
straight-through marking, left-turn marking, and right-turn marking.
5. Wrong-way driving
ITS cameras can detect the act of driving a motor vehicle against the direction of traffic.
An ITS camera continuously detects the moving direction of motor vehicles. When the
moving direction of a motor vehicle is opposite to the regulated vehicle direction on the
lane, the camera takes two or three snapshots (two snapshots by default), synthesizes the
snapshots into a violation image, and sends the image to the platform.
6. Unsafe lane change
ITS cameras can detect motor vehicles that change lanes illegally.
When a motor vehicle enters a surveillance lane, the camera takes a snapshot of the
vehicle and continuously monitors the movement path of the motor vehicle. If the
vehicle crosses the solid line and enters the adjacent lane, the camera takes the second
snapshot showing the lane change process. After the lane change, the camera takes the
third snapshot. The snapshots are synthesized into a violation image, which is then sent
to the platform.
7. Parking in yellow zones
ITS cameras can detect motor vehicles parked in yellow zones where the parking is
prohibited. The implementation process is as follows: The camera continuously monitors
the motor vehicles in its surveillance area. When a vehicle enters the yellow zone and
stays for a certain period (1–180s, configurable), the camera determines that a parking
violation occurs, takes two snapshots by default or three snapshots if specified, combines
the snapshots into a violation image and uploads the image to the platform.
8. Large vehicles driving on prohibited lanes

Huawei SDC
ITS cameras can detect large vehicles driving on lanes where large vehicles are
prohibited.
The camera continuously monitors a lane where large vehicles are prohibited. When a
large vehicle enters the lane, the camera determines that a violation occurs, takes two or
three snapshots (two snapshots by default), synthesizes the snapshots into a violation
image, and sends the image to the platform.
9. Bus lane violation
ITS cameras can detect motor vehicles in bus lanes where non-bus vehicles are
prohibited. The implementation process is as follows: The camera continuously monitors
the motor vehicles in its surveillance area. When a non-bus vehicle enters the bus lane
and stays for a certain period (1–180s, configurable), the camera determines that a
violation occurs, takes two snapshots by default or three snapshots if specified, combines
the snapshots into a violation image and uploads the image to the platform.
10. Motor vehicles driving non-motor vehicle lanes
ITS cameras can detect motor vehicles driving on non-motor vehicle lanes.
The camera continuously monitors the non-motor vehicle lane. When a motor vehicle
enters the lane and stays for a certain period (0–180s, configurable), the camera
determines that a violation occurs, takes 1–3 snapshots (two snapshots by default),
synthesizes the snapshots into a violation image, and sends the image to the platform.
11. Secondary vehicle feature recognition
ITS cameras deployed in checkpoint mode can perform secondary vehicle feature
recognition and take snapshots of in-vehicle violations (such as seat belt infractions and
hand-free device infractions) based on the secondary vehicle feature recognition.
When a vehicle passes, the ITS camera takes snapshots and performs secondary vehicle
feature recognition. If it detects that the driver or front passenger is not wearing the seat
belt or the driver uses handheld phones to make calls while driving, the camera
determines that a violation occurs, synthesizes the snapshots into a violation image, and
sends the image to the platform.
12. Emergency lane violation
ITS cameras can detect motor vehicles on emergency lanes.
The camera continuously monitors lanes. When a motor vehicle enters the emergency
lane, the camera determines that a violation occurs, takes 1–3 snapshots (two snapshots
by default), synthesizes the snapshots into a violation image, and sends the image to the
platform.
13. Illegal U-turn
ITS cameras can detect motor vehicles that make U-turns illegally.
The camera continuously monitors a motor vehicle lane. When a motor vehicle makes a
U-turn on the lane where U-turns are not allowed, the camera determines that a violation
occurs, takes three snapshots, synthesizes the snapshots into a violation image, and sends
the image to the platform.
14. Specified large vehicles that violate prohibitory traffic signs
ITS cameras can detect specified types of large vehicles driving on lanes where the large
vehicles are prohibited.

Huawei SDC

The camera continuously monitors a lane where specified types of large vehicles (the
large vehicle type is configurable) are prohibited. When a large vehicle enters the lane,
the camera determines that a violation occurs, takes snapshots, synthesizes the snapshots
into a violation image, and sends the image to the platform.
2.2.29 Vehicle Event Detection

Huawei SDCs can report alarms upon detecting some abnormal vehicle events and link alarm
linkage actions such as snapshot taking, video recording, and alarm output. Micro checkpoint
cameras support detection of non-motor vehicles driving on motor vehicle lanes, motor
vehicles driving on non-motor vehicle lanes, parking violation, wrong-way driving, illegal
backing, and marked lanes violation. Currently, ITS cameras support only detection of
non-motor vehicles driving on motor vehicle lanes. Other events are detected through the
traffic law enforcement process.
The vehicle event detection functions are as follows:
Non-motor vehicles driving on motor vehicle lanes: The camera can generate an alarm when
detecting a non-motor vehicle that drives on a motor vehicle lane. The detection sensitivity,
surveillance area, and alarm period are configurable.
Motor vehicles driving on non-motor vehicle lanes: The camera can generate an alarm when
detecting a motor vehicle that drives on a non-motor vehicle lane. The detection sensitivity,
surveillance area, and alarm period are configurable.
Marked lanes violation: The camera can generate an alarm when detecting a motor vehicle
that drives on the lane line. The detection sensitivity, surveillance area, and alarm period are
configurable.
Wrong-way driving/Illegal backing: The camera can generate an alarm when detecting a
motor vehicle that drives against the direction of traffic. The detection sensitivity, surveillance
area, and alarm period are configurable.
Parking violation: The camera can generate an alarm when detecting a motor vehicle that is
parked in a restricted place. The detection sensitivity, surveillance area, and alarm period are
configurable.
2.2.30 Smart Tracking

It always seems to be a contradiction between panoramic views and close-up images for video
surveillance. HD panoramic cameras can capture a panoramic image of a scene, but they
cannot obtain details. HD smart PTZ dome cameras can zoom in to capture image details, but
they cannot get a panoramic view of the scene. The smart tracking system can provide both
panoramic views and details at the same time. This system applies to large scenes such as
urban roads, urban high-altitude places, squares, harbors, urban crossroads, entrances and
exits of key organizations, road intersections, and campuses.
A box or bullet camera with a fixed focal length is usually used to get a panoramic view of a
scene.

Huawei SDC
Figure 2-32 Smart tracking
As shown in the preceding figure, the left image shows the live video image of a box or bullet
camera that is responsible for object detection and behavior analysis. When the detected
object triggers the behavior analysis rule, the coordinates of the box or bullet camera are
converted into the coordinates of the PTZ dome camera, and the PTZ dome camera is linked
to zoom in on the object. The right image shows the live video image of the PTZ dome
camera. After receiving instructions from the box or bullet camera, the PTZ dome camera
performs zoom and rotation. When the object moves, the box or bullet camera continues to
detect and track the object and deliver rotation instructions to the PTZ dome camera. In the
whole process, the box or bullet camera is responsible for object detection and tracking while
the PTZ dome camera is responsible for receiving instructions from the box or bullet camera
to perform zoom and rotation.
The procedure for implementing the smart tracking service is divided into the following steps:
Coordinate calibration for the box or bullet camera and PTZ dome camera:
The calibration process is to calculate the mapping between the coordinates of the box or
bullet camera and the coordinates of the PTZ dome camera. The input coordinates of the
obtained model are the coordinates (x,y) of the box or bullet camera, and the output
coordinates are the PTZ coordinates of the PTZ dome camera. There are two types of
calibration: automatic calibration and manual calibration.
The automatic calibration process is as follows: Adjust the rotation direction and zoom ratio
of the PTZ dome camera until the preview image of the PTZ dome camera coincides with that
of the box or bullet camera. Enable automatic calibration for the box or bullet camera. In this
case, the box or bullet camera will obtain the current frame of the PTZ dome camera and that
of the box or bullet camera, extract features from the two frames, and match the two images
by using the features. After the match, the box or bullet camera calculates the mapping model
between the box or bullet camera coordinates and the PTZ coordinates of the PTZ dome
camera based on the current PTZ information of the PTZ dome camera. The automatic
calibration is then complete.
The manual calibration process is as follows: Select a calibration point on the box or bullet
camera and adjust the PTZ device (zoom and rotation) of the PTZ dome camera to ensure that
the central point of the PTZ dome camera's preview image coincides with the calibration point
of the box (bullet) camera. Then, record the calibration point coordinates of the box or bullet
camera and the PTZ coordinates of the PTZ dome camera as a group of calibration points.
Repeat the preceding step to select 4–12 (six recommended) groups of calibration points.
Deliver the calibration points to the box or bullet camera for manual calibration on the
configuration page. In this case, the box or bullet camera calculates the mapping model
between the box (bullet) camera coordinates and the PTZ coordinates of the PTZ dome
camera based on the calibration points. The manual calibration is then complete.
Tracking policy configuration:

Huawei SDC
Smart tracking is triggered by behavior analysis. In typical application scenarios, it is used

together with intrusion detection, tripwire crossing detection, and area entry/exit detection. A
PTZ dome camera can track only one object at a time. If multiple objects trigger the behavior
analysis rule at the same time, you need to define a tracking priority policy. Currently, the
following two tracking policies are supported: 1. Continuously tracking the current object:
The PTZ dome camera continuously tracks the object that triggers the behavior analysis rule
until the object leaves the surveillance view. In this case, if another object triggers the
behavior analysis rule, the PTZ dome camera will track this object. 2. Tracking the current
object for a period of time: The PTZ dome camera tracks the object that triggers the behavior
analysis rule for a predefined period of time. When the predefined time elapses, the PTZ
dome camera will track the next object that triggers the behavior analysis rule. When there is
no object in the view, the PTZ dome camera automatically returns to the home position.
Smart tracking:
There are three application scenarios: automatic tracking, manual tracking, and manual
positioning.
Automatic tracking must be used together with behavior analysis. When an object triggers the
behavior analysis rule, the box or bullet camera instructs the PTZ dome camera to track the
object based on the tracking policy.
During automatic tracking, the box or bullet camera continuously detects objects in the view
and displays metadata on the live video image (displayed in the object frame). Users can
select an object in the live video image. In this case, the box or bullet camera stops automatic
tracking, continuously tracks the selected object until the object leaves the view, and then
restores to the automatic tracking mode. During manual tracking, users can manually select an
object to track at any time.
Manual positioning is similar to 3D magnification and positioning on a PTZ dome camera.
Users need to perform operations on the live video image of the box or bullet camera. Then
the box or bullet camera instructs the PTZ dome camera to perform 3D magnification and
positioning, while the box or bullet camera continuously monitors the panoramic view.
Manual positioning and tracking (including automatic tracking and manual tracking) are
mutually exclusive. That is, you need to disable the tracking function before using the manual
positioning function.
SDK interfaces are provided for smart tracking configuration and tracking. The web mode is
not supported. Currently, the SiMS has integrated the configuration and tracking interfaces.
The VCN platform has integrated only the tracking interface. The third-party platform or
client can integrate the interfaces based on specific requirements.
2.2.31 1 + N (Master-Slave Camera Surveillance)

The long-term development of video surveillance systems leads to a large number of cameras
on the live network. During old network reconstruction, how to make full use of legacy device
capabilities to reduce network reconstruction costs is mission-critical task in the current video
surveillance system design. The 1 + N feature is a solution to this scenario. A smart camera
that supports the 1 + N feature, functioning as the master camera, can not only process its own
services but also perform video stream connection, decoding, intelligent analysis, and
intelligent analysis result output for one or more common cameras through access protocols.
In this way, the common cameras can obtain some intelligent analysis capabilities.
The 1 + N function applies to multiple scenarios such as lanes where motor vehicles,
non-motor vehicles, and pedestrians may appear at the same time as well as crossroads.

Huawei SDC
2.2.32 Multiple-Algorithm Concurrency

Multi-algorithm concurrency refers to that a camera can run multiple algorithms at the same
time, and each algorithm implements only a single service function. For example, face
detection, facial recognition, facial attribute extraction, full-body capture, or motor vehicle
capture is considered as a single service. Multiple services form an algorithm mode (for
example, face detection, facial recognition, and facial attribute extraction form the face
capture mode). Multiple algorithms can run concurrently in one algorithm mode. A camera
can switch among multiple algorithm modes.
The algorithm modes that support concurrent operation of multiple algorithms include face
checkpoint, micro checkpoint, full intelligence, and behavior analysis modes.
1. In face checkpoint mode, the following algorithms are supported: face capture, facial
attribute extraction, facial feature extraction, full-body capture, personal attribute
extraction, and people flow statistics.
2. In micro checkpoint mode, the following algorithms are supported: license plate capture,
license plate recognition, vehicle capture, vehicle attribute extraction, full-body capture,
personal attribute extraction, non-motor vehicle capture, non-motor vehicle attribute
extraction, vehicle flow statistics, and traffic violation detection (abnormal license plate,
non-motor vehicles driving on motor vehicle lanes, motor vehicles driving on non-motor
vehicle lanes, wrong-way driving, marked lanes violation, and illegal backing)
3. In full intelligence mode, the following algorithms are supported: face capture, facial
attribute extraction, full-body capture, personal attribute extraction, license plate capture,
license plate recognition, vehicle attribute extraction, non-motor vehicle capture, and
non-motor vehicle attribute extraction
4. In behavior analysis mode, the following algorithms are supported: crowd situation
analysis (head counting, crowd density detection, crowd gathering/dispersing detection,
and queue length detection), full-body capture, personal attribute extraction, and
behavior analysis (tripwire crossing detection, intrusion detection, area entry/exit
detection, loitering detection, fast movement detection, abandoned object detection,
removed object detection, and lens blocking detection)
2.2.33 Heat Map

The heat map function is applicable to indoor scenarios such as shops, supermarkets,
museums, and shopping malls. Through video surveillance, the camera can collect the heat
information of different areas in different periods, and generates daily, weekly, monthly, or
annually heat maps and statistical reports, providing key basis for resource management and
management decision-making. For example, in a large shopping mall, if the historical
customer flow data can be analyzed to obtain the crowd density data of each area in different
time segments, the management personnel of the shopping mall can know the customer
preference. Then they can adjust commodities and advertisements in an area in different time
segments to increase the business income.

Huawei SDC
2.2.34 MAC Address Collection

The MAC address collection function is mainly used in finance, telecom, government, school,
airport, factory, hotel, museum, and traffic surveillance scenarios that pose high requirements
on HD video quality but has dimmed light or no light. The camera can proactively collect
terminal data such as wireless hotspots and mobile phone MAC addresses and periodically
upload the data to the police platform.
For the police, the Wi-Fi probe technology is used to obtain MAC addresses and related data
of terminal devices, which assists the police in case investigation, selected personnel tracking,
and mass incident evaluation and warning.
2.2.35 T-Shot Intelligent Snapshot Engine

The T-Shot intelligent snapshot engine uses the framing policy to perform pedestrian
detection and license plate detection at the same time. When only one sensor is available, the
T-Shot engine can ensure proper exposure on objects, especially faces and license plates, in
different scenarios. Faces are easy to get underexposed at night. Therefore, the shutter speed
needs to be decreased and the gain needs to be increased to clearly capture faces. License
plates are easy to get overexposed at night. Therefore, the shutter speed needs to be increased
and the gain needs to be decreased to clearly capture license plates. The T-Shot engine uses
the multi-channel exposure principle. That is, the previous frame is used for face capture, and
the next frame is used for license plate capture. The face capture video stream improves the
image brightness to ensure that the face details are clear. The license plate capture video
stream lowers the image brightness to ensure that the license plate characters are clear. The
T-Shot engine helps clearly capture both face images and license plate images in the same
scenario. It is mainly used in parking lots, parks, scenic spots, crossroads, and other places
with high vehicle and people flows.
2.2.36 Traffic Data Collection

ITS cameras can collect vehicle traffic statistics.
Statistical data:

Huawei SDC
In ePolice or checkpoint scenarios, the camera can collect statistics on the traffic volume,
vehicle type, vehicle direction, average speed, queue length, time headway, space headway,
lane time occupancy, lane space occupancy, and traffic status by lane and period.
2.2.37 Dual-Frame Integration

The dual-frame integration implements high-speed switchover between the infrared frame and
the common frame. The working principle is as follows: A dual-passband filter and an
infrared lamp are used to generate common frames and infrared frames alternately (infrared
lamp enabled: infrared frame generated; infrared lamp disabled: common frame generated).
The infrared frame provides object details, and the common frame provides object colors. The
DSP integrates common frames and infrared frames to product color images in low-light
conditions.
Before integration After integration
2.3 Security
2.3.1 Digital Watermark and Media Security Technologies
With wide application of codec formats that comply with international standards and
ever-developing audio and video processing technologies in the video surveillance industry,
video processing applications are relatively mature. However, re-coding and video data
tampering cases frequently occur. It is an indispensable capability of the video surveillance
solution to ensure data transmission security and data integrity. Focusing on video encoding
universality and considering industry appeals, Huawei embeds digital watermark and media
security technologies into cameras. This effectively prevents audio and video data from being
tampered with, ensures data integrity and authenticity, prevents network cracking, and
enhances media data security.
1. Digital watermark technology
During stream data output through video encoding, watermark information (function as
protection information) related to the stream frame (including the current number of
frame bytes, time, device MAC address, and SN) is stored in the stream data packet

Huawei SDC
defined by users. The data packet is stored in disks or transmitted with the compressed
stream.
Figure 2-33 Watermark encryption at the encoding end

Media stream
storage
Feature processing: provides the
current number of frame bytes, time,
device MAC address, and SN.
Original
image Digital image RTP media stream Streaming
encoding media unit
Watermark encryption: combines

features with media streams.
Streaming media
forwarding
The watermark cannot be perceived, does not affect protected data usage, and will not degrade the image
quality.
At the media playing end, the stream data packet is decoded to obtain the features related
to the watermark, including the current number of frame bytes and time and compare the
features with the preset video encoding information. At the application layer, the system
can check whether the decoding output information is the same as the protection
information of the data packet to determine the stream validity. The system verifies the
watermark information to ensure data security and integrity. If the watermark fails to be
verified, the system generates an alarm.
Figure 2-34 Watermark verification at the decoding end

Feature processing: extracts
watermark information from the Alarm information
stream data packet. re
ailu output
on f
r i fi cati
Ve
Video decoding end
Verific
ation
su cce Normal decoding
ss
Watermark verification: verifies

watermark information in the data packet.
Video file with watermark encryption
2. Media security technology

With the media security technology, the Huawei video surveillance platform generates
initial encryption information and sends the encrypted information to cameras. The
cameras obtain the final private key based on negotiated encryption rule, encrypt the
collected video data and final private key in combination mode, and send the encrypted
video data to the surveillance center. After receiving encrypted streams, the platform
decrypts the private key, preventing media information theft and enhancing media
transmission security.

Huawei SDC
Figure 2-35 Secure decryption of online media streams

The platform sends initial
encryption information. Generate the initial
encryption information
Obtain the final private key.
Original RTP encrypted

image Digital image media stream Video surveillance
Stream output
encoding platform
Media encryption:
Media
video stream +
decryption
private key
2.3.2 Video Buffering

Video and signaling data of cameras is transmitted over networks, so network reliability is an
important reference for measuring the reliability of a video surveillance system. However,
network outages and faults are inevitable in actual video surveillance systems. How to ensure
data integrity during network outages is an issue that a field-proven video surveillance
solution must cope with.
With video buffering technology, video data can be temporarily stored in the buffer and SD
card of a camera upon a network outage, preventing video data loss and ensuring data
integrity during network outages.
Figure 2-36 Video buffering technology
Actual network
outage point
Normal upload Network Normal upload
outage
of data detection
of data
point
Network
recovery point
Data stored in the local Video data stored in

buffer area before the local SD card after
network outage network outage
The central platform

checks data integrity
If an exception occurs on the network between a camera and the IVS platform, the camera can
detect the exception and enable the video buffering function. Then, video data transmission is
changed from network transmission to local storage. The video data is stored in the SD card
of the camera. If the network outage lasts for a long period of time and the size of recorded
video exceeds the preset upper limit, data is cyclically overwritten in the SD card.
When the network is recovered, the central platform initiates a recording downloading request,
automatically searches for absent recording time segments based on the predefined recording
plan, and downloads absent recordings from the camera, ensuring video data integrity and
continuity. After the absent recordings are downloaded from the camera, the camera
automatically deletes the buffered data.
To implement the video buffering function, cameras must support this function, have an SD card, and
cooperate with the central platform.

Huawei SDC
2.3.3 Video Buffering Enhancement

After the platform goes offline, the device records the alarm information and alarm-triggered
recording generated during the offline period. After the platform goes online, you can query
the alarm information and alarm-triggered recording generated during the offline period.
When the link is disconnected, the alarm-triggered recording can be stored in the SD card and
the alarm information is recorded. The alarms include motion detection alarms, lens blocking
alarms, Boolean value input alarms, and intelligent analysis alarms.
When the link recovers, the device can respond to the alarm information and alarm-triggered
recording requests from the platform.
2.3.4 IP Address and MAC Address Filtering

Video data of cameras may be illegally accessed by unauthorized personnel. As a result,
information leakage may occur. To protect video data, cameras must be able to filter out IP
data packets by IP address, port number, MAC address, and transport layer protocol.
The implementation principle is to integrate the IP packet filtering system integrated by Linux
kernel hardened by Huawei to better control IP packet filtering, MAC address filtering, and
firewall configurations.
Huawei SDCs allow users to flexibly set data packet filtering rules, similar to the functions of
a network firewall, to filter out data packets by specific IP addresses, port numbers, MAC
addresses, or transport layer protocols.
1. After a blacklist is configured, cameras deny the data packets that meet the rule specified
in the backlist.
2. After a whitelist is configured, cameras allow the data packets that meet the rule
specified in the whitelist.
3. Multiple rules can be configured. The configured rules can be queried, modified, and
deleted.
Figure 2-37 IP address filtering
2.3.5 IEEE 802.1X Access

In the absence of access authentication, unauthorized users and devices can easily connect to
the video surveillance network, causing information leakage or even system crash.
The Huawei network video surveillance system uses 802.1X access authentication. Cameras
support 802.1X access authentication (specifically, EAP-TLS authentication) as requesting

Huawei SDC
parties. Users can configure required authentication information on web clients. After 802.1X
access authentication is enabled, malicious attacks at the access layer can be prevented. This
provides access authentication security for the surveillance system and prevents unauthorized
users or devices from accessing and attacking the video surveillance system.
Figure 2-38 IEEE 802.1X access
Unauthorized device
Surveillance
data center
Unauthorized
user
Network
(802.1X access)
Camera
(802.1X access)
2.3.6 TCP Congestion Control

Network congestion may occur if video data packets that cameras send using TCP exceed the
network's bearer capability. When network congestion occurs, packets will be dropped,
causing artifacts and frame freezing.
Huawei SDCs use Huawei's proprietary TCP congestion control algorithm to determine
whether network congestion occurs based on TCP packet loss conditions. If network
congestion occurs, the cameras select appropriate congestion control policies based on
network congestion conditions to reduce the transmission rate. This helps mitigate network
congestion and therefore ensures smooth video.
TCP congestion control takes effect only when the transmission protocol is RTP over TCP.
2.3.7 KMC
The key management CBB (KMC) is a Huawei-proprietary security management mechanism
that provides secure storage and lifecycle management capabilities for keys used in products
and services. The KMC solves many security issues of the product, such as using insecure
encryption algorithms, incorrectly using parameters of security algorithms, non-standard or
lack of key management, and lack of key management components that support multi-device
and multi-process.
The KMC supports the following functions:
 Hierarchical key management for security isolation
 Role-based key management
 Key lifecycle management
 On-demand key set expansion, and key isolation between different apps
 Keys stored in files in active/standby mode, as well as key import and export

Huawei SDC
2.3.8 Security Certification

Huawei passed the security evaluation of the independent evaluation institution Brightsight in
Europe in April 2018 and becomes the world's first camera vendor that obtains the Common
Criteria (CC) EAL3+ certificate. This certificate proves that Huawei's camera products take
the lead in terms of security and stability across the globe and can provide customers with
secure and reliable services.
Rapid network development causes severe challenges to cyber security. Therefore, more and
more customers pose higher requirements on security. Huawei makes it an important
development strategy to build an end-to-end global cyber security assurance system. In the
Statement on Establishing a Global Cyber Security Assurance System, Huawei emphasizes
that "our commitment to cyber security will never be outweighed by the consideration of
commercial interests."
As a pioneer in the Safe City video surveillance solution, Huawei SDCs are equipped with
industry-leading encryption algorithms such as AES256, comprehensive log audit functions,
security isolation design, and hierarchical key permission management. This provides robust
security protection for customer data from four aspects: network architecture, algorithm
application, security management, and camera system.
2.3.9 GB 35114
GB 35114 is the Technical Requirements for Information Security of Video Surveillance
Network System for Public Security formulated by the Ministry of Public Security of the
People's Republic of China. It was released in November 2017 and forcibly implemented
from November 2018. The GB 35114 protocol enhances the security capability based on the
GB/T 28181 protocol. The core requirements of this specification include device (client,
network camera, and third-party platform) two-factor authentication, video encryption, and
video signature, aiming to prevent unauthorized device access and protect video content
security (protection against content tampering and leak). In addition, this specification
proposes security requirements such as message authentication (message integrity), video
export, permission management, and log management.

Huawei SDC
Huawei network cameras support the GB 35114 protocol that provides the following
capabilities:
1. Two-factor authentication based on the digital certificate and management platform
2. Signaling signature, preventing signaling from being tampered with
3. User identity authentication, allowing only authorized users to log in to the system
2.3.10 Intrusion Detection

In recent years, the rapid development of IP networks promotes the video surveillance
industry to enter the all-network era. However, the networking feature of the network cameras
also causes frequent occurrences of security vulnerabilities.
The intrusion detection component provides the following functions: promiscuous network
adapter detection, key file tampering detection, unauthorized super user detection, botnet
detection, rootkit detection, and CoinMiner virus detection. This component performs security
detection on the networks, files, users, and malicious programs of cameras. An intrusion
detection system is installed on a camera to provide multi-layer visualized intrusion detection
capabilities. This helps detect hacker intrusion behavior in a timely manner and protect
camera security. Currently, no other enterprise in the security industry has deployed an
intrusion detection system on cameras. Huawei is the first vendor in the industry to embed
intrusion detection capabilities in cameras.
2.4 Reliability
2.4.1 Three Systems and Three Configurations
Traditionally, if the system configuration of a camera is invalid or lost due to factors such as
power outage during configuration change, engineers need to climb the pole to remove it for
repair, which causes long service interruption and high recovery cost. The systems of Huawei
SDCs have three configurations, two in 1:1 image backup and one read-only default
configuration. When the configuration file being written is damaged because of, for example,
power outage, the other file of the two in 1:1 image backup can automatically recover data.
When the two files are both damaged (with a very low probability), the system enables the
default configuration file to ensure proper system running. Engineers do not need to climb the
pole to remove related cameras for repair.
2.4.2 Wide Temperature and Voltage Ranges

Huawei SDCs can function properly within a wide temperature range (–40°C to +60°C). They
are also tolerant of voltage fluctuations (12 V DC (±25%), 24 V DC (±25%), and 24 V AC
(±24.9%)) and can run properly without unexpected restart or failure in such environments.
The cameras also support both adapter and PoE power supplies in mutual hot backup mode.
2.4.3 Ingress Protection Rating

Ingress protection (IP) rating classifies and rates the degrees of protection provided against
intrusion (body parts such as hands and fingers), dust, accidental contact, and water by
mechanical casings and electrical enclosures. Huawei outdoor-mounted cameras support at
least IP66. The flagship camera models support IP67. An IP rating consists of two digits.

Huawei SDC
The first digit indicates the level of protection that the enclosure provides against access to
hazardous parts (for example, electrical conductors and moving parts) and the ingress of solid
foreign objects. A larger digit indicates a higher protection level.
The second digit indicates the level of protection that the enclosure provides against harmful
ingress of water. A larger digit indicates a higher protection level.
First Digit Object Size Description

Protected Against
0 No protection No protection against contact and ingress of

objects.
1 > 50 mm Protection from a large part of the human body

such as a hand (but no protection from
deliberate access); protection from solid objects
greater than 50 mm in diameter.
2 > 12 mm Protection against fingers or other objects

greater than 12 mm in diameter.
3 > 2.5 mm Protection from entry by tools, wires, and other

objects greater than 2.5 mm in diameter.
4 > 1 mm Protection from entry by tools, wires, and other

objects greater than 1 mm in diameter.
5 Dust protected Ingress of dust is not entirely prevented, but it

must not enter in sufficient quantity to interfere
with the satisfactory operation of the equipment;
complete protection against contact (dust proof).
6 Dust tight No ingress of dust; complete protection against

contact (dust tight).
Second Digit Protected Against Description
0 No protection No protection against water drops.
1 Dripping water Dripping water (vertically falling drops) shall

have no harmful effect.
2 Dripping water Vertically dripping water shall have no harmful

when tilted up to effect when the enclosure is tilted at an angle up
15° to 15° from its normal position.
3 Spraying water Water falling as a spray at an angle up to 60°

from the vertical shall have no harmful effect.
4 Splashing of water Water splashing against the enclosure from any

direction shall have no harmful effect.
5 Water jets Water projected by a nozzle against the

enclosure from any direction shall have no

Huawei SDC
Second Digit Protected Against Description

harmful effect.
6 Powerful water jets Water projected in powerful jets against the

enclosure from any direction shall have no
harmful effect.
7 Immersion up to 1 Ingress of water in a harmful quantity shall not

m be possible when the enclosure is immersed in
water under defined conditions of pressure and
time (up to 1 m of submersion).
8 Immersion beyond The equipment is suitable for continuous

1m immersion in water under conditions which
shall be specified by the manufacturer.
Normally, this will mean that the equipment is
hermetically sealed. However, with certain
types of equipment, it can mean that water can
enter but only in such a manner that it produces
no harmful effect.
2.4.4 IK10 Impact Protection Rating

IK classifies the degrees of protection provided by enclosures against external mechanical
impacts when the rated voltage of the protected equipment is not greater than 72.5 kV. An IK
rating is determined by the impact of a free fall hammer in unit of joule. IK10 can withstand
20J impact.
IK01 IK02 IK03 IK04 IK05 IK06 IK07 IK08 IK09 IK10
0.14J 0.2J 0.35J 0.5J 0.7J 1J 2J 5J 10J 20J
2.4.5 Class D Anti-Corrosion

In the natural environment, the main factors causing device corrosion are listed as follows:
salinity, temperature, sunlight, and humidity. A seaside scene may be the harshest for security
cameras. Based on the experience in outdoor application of wireless equipment, Huawei
proposes a set of salt corrosion standards, classifying environment conditions into different
levels based on the corrosion factors. A Class D environment is most likely to cause severe
corrosion. Huawei sets up an experiment field at Wanning, Hainan, where all Huawei
products can be installed to test their anti-corrosion capabilities. Huawei cameras meeting the
Class D anti-corrosion requirements can be used within 1500 meters from the sea or on an
offshore platform. The cameras are free from blistering, cracking, or corrosion within three
years of use. No functionality or water leakage problems will occur within five years. These
cameras can resist deterioration from a 30-day salt spray test in accordance with IEC
60068-2-11.
Class D anti-corrosion camera models are IPC6681-Z20-Y, IPC6285-VRZ, and IPC6285-VWZ-LPR.

Huawei SDC
2.4.6 Dual System Backup

To ensure the version stability and reliability, two copies of the following key data are backed
up in the flash memory of the camera: boot data, version files, and configuration files. Besides,
the basic software package and the algorithm package are separated from each other.
2.5 Ease of Use

2.5.1 ABF
With the auto back focus (ABF) function, the relative position of the CCD/CMOS is adjusted
to deliver clear images. Cameras that have been used for a long time are prone to be out of
focus due to external forces such as vibration, requiring frequent focus adjustment. ABF
makes focus adjustment more convenient and accurate.
Figure 2-39 ABF
Lens
2.5.2 OEC
Opto-electronic cascade (OEC) applies to the surveillance site with two box cameras or one
box camera linked with one PTZ dome camera. The following solutions are available in this
scenario:
Solution 1: two optical fibers, one PTZ dome camera with optical fibers, and one box camera
with optical fibers

Huawei SDC
Figure 2-40 Solution 1
Gate
Optical fiber
High device costs

High construction costs
Solution 2: one optical fiber, one switch equipped with an optical module, one common PTZ
dome camera, and one common box camera
Gate
Optical fiber
Switch
High device costs
Large space occupation
Solution 3: OEC (one optical fiber, one camera with an OEC interface, and one common
camera)

Huawei SDC
Gate
Optical fiber
Low device costs

Low construction costs
Space occupation not increased
2.5.3 GPON Access

Compared with EPON schemes of other vendors, the GPON has the following technical
features: (1) Orientation to the telecom operation access network (2) High bandwidth: The
downlink rate can reach 2.488 Gbit/s and the uplink rate can reach 1.244 Gbit/s. (3) High
transmission efficiency: The downlink transmission efficiency can reach 94% (the actual
bandwidth reaches 2.4 Gbit/s) and the uplink transmission efficiency can reach 93% (the
actual bandwidth reaches 1.1 Gbit/s). (4) Full support for services: According to the G.984.X
standard, all telecom-level services (including voice, data, and video services) are supported.
(5) Strong management capability: Rich functions are available, sufficient OAM domains are
reserved in the frame structure, and the OMCI standard is formulated. (6) High QoS: There
are multiple QoS levels to meet service bandwidth and delay requirements. (7) Low
comprehensive costs: The transmission distance is long and the split ratio is high. As a result,
OLT costs are allocated and user access costs are reduced effectively.
Installation scenario: The single-fiber bidirectional transmission mode is used, the
transmission distance can reach 20 km, and optical splitters are used for connection. This
mode is applicable to a safety project where multiple devices are connected by a single cable,
saving optical fiber resources.
2.5.4 Openness Design

To ensure camera interoperability between different systems, cameras need to provide
multiple interface protocols to adapt to different system environments. Huawei SDCs provide
multiple standard protocols for information exchange between network video devices. With
different types of standard interfaces provided, such cameras are highly open-ended and easy
to use, and can be connected not only to Huawei's video surveillance platform but also to
platforms of other mainstream vendors.
1. TEYES protocol
Huawei SDCs can be connected to Huawei's video surveillance platform using China
Mobile's TEYES protocol and provide a wide assortment of services for customers, such
as live video surveillance, recording search and playback, PTZ controls, alarm detection
and linkage, voice intercom, and electronic map.
2. Software development kit (SDK)
Huawei SDCs provide an SDK, which offers friendly secondary development interfaces
for open platforms. Such interfaces help improve Huawei SDCs' openness and

Huawei SDC
applicability, as the interface definition meets user habits and allows smooth IPC access.
Huawei SDCs provide two SDK versions, Windows and Linux, for implementing
functions such as parameter setting, user login (including registration keep-alive), PTZ
controls, real-time traffic diversion (including audio and video streams), voice intercom
traffic diversion, and alarm reporting.
3. ONVIF protocol
Huawei SDCs support the ONVIF protocol (ONVIF 2.1, ONVIF 2.2, or ONVIF Profile
S) for interconnecting with network video products of different vendors, maximizing
customers' return on investment (ROI). Using this protocol, Huawei SDCs support a
variety of functions such as device management, device discovery, image configuration,
device input and output service, media configuration, PTZ controls, real-time streaming
media, and event processing.
4. GB/T 28181 protocol
Huawei SDCs support the GB/T 28181 protocol. Users can connect such cameras to
Huawei's video surveillance platform or a third-party platform using the GB/T 28181
protocol to implement services such as live video viewing, PTZ controls, and alarm
reporting.
2.5.5 Adaptive Rapid Deployment

During deployment and application, the increasing scale of the video surveillance system
requires more costs on camera configuration, installation, commissioning, and maintenance.
During camera deployment, users need to configure camera network parameters and
parameters of the video surveillance platform to be connected. In many situations, users also
need to consider the relationship between cameras and the corresponding video surveillance
platform. To ensure correct parameter settings, users need to have considerable expertise in
cameras and video surveillance but in fact, most of them do not acquire such knowledge.
The Huawei video surveillance solution uses adaptive rapid deployment technology and
centrally deploys the DHCP server on the surveillance platform for automatic registration and
plug-and play of cameras. This helps effectively reduce installation and deployment costs as
well as improve camera installation and maintenance efficiency.
Figure 2-43 Automatic camera discovery
Detecting new devices
SVN:20:12:07:02:20:57
Model: eSpace IPC 5811-
WD-Z20
Vendor: Huawei
View
1. Set the IP address obtaining mode of a camera to DHCP. Then, the camera searches for
the DHCP server using DHCP in the network. When receiving the search information
from the camera, the DHCP server sends a response telling the camera that it has

Huawei SDC
received the search information and provides the DHCP server connection mode. The
camera applies to the server for an IP address according to the connection mode provided
by the DHCP server. The DHCP server automatically allocates an IP address to the
camera using UDP when receiving the request. The camera obtains this IP address and
also the IP address and port number of the surveillance platform.
2. The camera then sends registration information to the platform according to the
platform's IP address and port number, and obtains its own registration ID, which
uniquely identifies the camera. The automatic camera registration is then completed.
2.5.6 Multiple Power Supply Modes

Box cameras generally support 12 V DC power supply, while PTZ dome cameras support 24
V AC. However, in actual projects, the mains supply may be unavailable and only 24 V AC or
24 V DC is available, or the solar power supply is used, failing to meet power supply
requirements of box cameras and PTZ dome cameras.
Huawei SDCs support multiple power supply modes: 12 V DC, 24 V DC, 24 V AC, and PoE
for box cameras, and 24 V DC, 24 V AC, and non-standard PoE (–56 V to –57 V) for PTZ
dome cameras. When the mains supply is not available in a project, different types of Huawei
SDCs can use a uniform power supply that is available onsite, for example, 24 V DC or 24 V
AC. When the solar power supply is used, because Huawei SDCs support different voltages,
allowing for flexible power output of the solar power device, the solution adaptability can be
improved and costs can be reduced.
Additionally, Huawei SDCs are insensitive to positive and negative poles of the DC power
supply to avoid device damage due to incorrect connection of the positive and negative poles.
2.5.7 PoE, PoE+, and PoE++

Professional electricians with employment licenses are required to perform onsite deployment
of common AC power supply equipment. To reduce device deployment costs and improve
deployment convenience, the PoE technology is developed to concurrently supply power and
transmit data through existing Ethernet cables. Therefore, applications do not need to be close
to the AC power supply, so that common engineers can also complete the onsite deployment.
In 2003, the PoE specification, known as 802.3af, was released, which allowed the 48 V DC
power supply to provide a maximum power of 13 W. This specification was initially well
accepted. However, the upper limit 13 W restricted the number of applications. In 2009, IEEE
released a new standard called 802.3at or PoE+, which had higher voltage and current
requirements so that a maximum power of 25.5 W can be provided.
Currently, all Huawei SDCs support IEEE 802.3af/at PoE.
With the enhancement of camera functions (infrared night vision, low-temperature heating,
and active defogging), there are higher power requirements. The maximum power of a PTZ
dome camera reaches 50 W, which is far beyond the maximum (25.5 W) specified in PoE+.
With its extensive experience in the PoE field and advantages of the overall solution, Huawei
launched the high-power PoE solution based on the PoE++ technology, which provides a
convenient power supply mode for high-power cameras.
Compared with the power supply solution in which standard PoE equipment uses four wires
in the network cable for power supply, Huawei PoE++ solution uses all the eight wires in the
network cable to increase the through-current capability, so as to increase the maximum PoE
power and reduce line loss. Either a Huawei PoE++ switch or a Huawei PoE++ power adapter
can be selected for a PoE++ camera.

Huawei SDC
2.5.8 Long-Distance Ethernet Transmission

The common network transmission distance is about 100 m with little tolerance of
temperature change, network cable quality, and other factors. Some Huawei SDC models
support long-distance Ethernet transmission. At 100 Mbit/s, data can be transmitted about 200
meters through Ethernet.
The secrete to longer transmission distance is better quality of signals transmitted on the
network. The Marvell industrial-grade PHY chip is used to ensure that the amplitude of the
signal to be transmitted meets the long distance transmission requirement. Additionally, the
chip can automatically calculate the signal-to-noise ratio (SNR). This is implemented by
sending test signals and determining the signal quality of the entire link based on the loopback
function. This function requires the support from the peer network switches. Currently, the
cameras can be used with the following Huawei network switches: S5720I-12X-SI-AC,
S5720I-12X-PWH-SI-DC, S5720I-28X-SI-AC, and S5720I-28X-PWH-SI-AC.
Category 5e network cables must be used. Otherwise, the transmission distance will be affected.
Figure 2-44 Long-distance Ethernet transmission
2.5.9 Automatic Preventive Maintenance and Fault Reporting

Regarding the video surveillance system, operations are not limited to services such as live
video viewing and video storage. When a camera is faulty, locating the faulty camera among a
large number of cameras is time-consuming, which increases system O&M costs. Therefore,
how to quickly and accurately locate a faulty camera is necessary in routine maintenance of
the video surveillance system. As the core member in Peripheral Units (PUs), Huawei SDCs
use automatic preventive maintenance and fault reporting design for detecting camera system
running status (such as temperature and SD card reading and writing) in polling mode. This
improves maintenance efficiency and reduces maintenance costs.
1. Temperature alarm
Huawei SDCs provide a built-in temperature sensor to monitor the ambient temperature
inside the camera in real time. When the sensor detects that the camera has been running
at a temperature beyond the upper limit within a period, the camera automatically
generates an alarm and reports the alarm to the surveillance center. When the
surveillance center receives the alarm together with the camera information, maintenance

Huawei SDC
personnel can be sent to that camera in a timely manner for troubleshooting. This helps
avoid device damage and ensure surveillance quality.
2. Automatic temperature adjustment (PTZ dome camera)
Huawei PTZ dome cameras support automatic temperature adjustment. When the
temperature inside a camera is higher or lower than the temperature for secure running,
the camera automatically enables the fan or heater through the internal logic circuit for
temperature adjustment. If the temperature adjustment is proved to be ineffective, the
camera automatically generates an alarm and reports the alarm to the surveillance center,
requesting manual maintenance services.
3. SD card fault detection
Huawei SDCs can detect SD card reading and writing faults and generate alarms. When
an SD card is inserted into the card slot, the camera enables SD card mounting detection.
When detecting an SD card reading and writing exception, the camera automatically
generates an alarm and sends the alarm to the surveillance center.
When the SD card being used for data writing and reading encounters exceptions, the
camera also generates an alarm, records the alarm to the alarm log, and reports the alarm
to the surveillance center.
2.5.10 Remote Zoom and Focus Control

Cameras support zoom and focus adjustment either in manual or motorized mode. In manual
mode, users can adjust the zoom ring on the lens to select a field of view for the surveillance
area. However, after the camera is installed and fixed in a position, it is inconvenient to
manually adjust the zoom ring frequently.
Huawei SDCs support remote zoom and focus adjustment on web clients. This function
further simplifies installation and commissioning and improves product flexibility.
Figure 2-45 Remote zoom and focus control
Sends Zoom/Focus control signal
IP IP
Client
Camera Switch
Zoom + Focus
Zoom/Focus control
2.5.11 G-sensor Image Stabilization

To prevent images produced by cameras installed on poles from becoming blurry due to
factors such as wind blowing, bridge quaking, and large vehicle passing and improve image
clarity, Huawei rolls out the G-sensor image stabilization technology.
Image stabilization technologies include electronic image stabilization and optical image
stabilization technologies. Electronic image stabilization can correct image shaking using
algorithms, which is mainly used in the video surveillance industry. Optical image
stabilization can implement physical shaking supplementation by adding supplementary

Huawei SDC
components to optical paths. The optical image stabilization technology is mostly used in the
consumer electronics field.
Huawei uses efficient stabilization technologies, that is, a combination of the G-sensor
(electronic gyroscope and accelerometer) and a stabilization algorithm, to present a more
effective and accurate stabilization feature.
Working principle:
Electronic image stabilization assisted by the G-sensor
Figure 2-46 G-sensor working principle
Optical Compensa- Image

G-sensor
axis angle tion stablization
Device jitter
Original
Camera Image data
image
Electronic image stabilization is a technology that completely uses algorithms to estimate the
shaking and implement shaking supplementation. A surveillance camera obtains the estimated
motion vector by comparing the difference between the image corresponding to the current
frame and that corresponding to the previous frame, uses the motion vector as a mapping
matrix, and calculates the mapping result of the current frame. In this way, the image after
shaking correction is obtained.
The effect of electronic image stabilization depends on the accuracy of motion vector
estimation. If the motion vector is inaccurately estimated, no accurate correction value can be
obtained later. In a traditional solution, motion vectors are estimated based only on motion
changes of images. The accuracy and response time are not ideal. In addition, estimation
deviation is caused by local motion and motion blur.
Huawei uses the G-sensor (electronic gyroscope and accelerometer) to replace the motion
estimation part in the stabilization algorithm, which samples gyroscope data in real time to
obtain the current three-axis (X/Y/Z) space posture angle. The horizontal and vertical offsets
of the image relative to the initial state can be obtained based on the posture angle and focal
length. These offsets are used to implement image correction.
2.5.12 3D Positioning of High-Speed PTZ Dome Cameras

With continuous development of applications in the video surveillance industry, users have
increasingly higher requirements on image clarity and remote control. To address such
requirements, Huawei provides not only the entire HD video surveillance solution but also 3D
positioning technology in the PTZ dome camera solution based on the existing PTZ
communication control.
1. When users need to check details in a scenario in the live video, what they need to do is
to draw a dotted box in the live video. The position of the dotted box in the screen and
the full-screen aspect ratio are then calculated, and the control signal is sent to the PTZ
dome camera through the TEYES protocol.
2. The PTZ dome camera adjusts the zoom-in ratio according to the control signal, quickly
locates the area corresponding to the selected box by rotating and zooming, and displays
this area in full screen mode (as shown in Figure 2-47). PTZ controls using 3D

Huawei SDC
positioning allow users to quickly obtain the image they want, improving surveillance
efficiency.
Figure 2-47 3D positioning effect of a PTZ dome camera
Center positioning Width
Selected box
Height
Center positioning
3D positioning can be implemented on Huawei PTZ dome cameras only when Huawei IVS software is
used.
2.5.13 Stream Profiles

Cameras provide default settings for all stream parameters before delivery, but the default
settings are not suitable for varying surveillance scenarios. In this case, stream parameters
need to be adjusted based on actual conditions, which is time-consuming.
Huawei SDCs have pre-defined four stream profiles: high quality, balanced, low bandwidth,
and mobile devices. Besides, users can customize at most four profiles. Users can select a
similar stream profile and adjust only a few parameters in that profile based on their actual
conditions, sharply improving camera deployment efficiency and alleviating deployment
workload.
1. High quality: This stream profile ensures image quality with preference and applies to
scenarios with sufficient bandwidth. In this case, the primary stream bandwidth may
reach 8 Mbit/s, and each camera may require more than 20 Mbit/s by also considering
dual-channel traffic diversion and video buffering. When only single-channel traffic
diversion is required and video buffering is not, the recommended bandwidth is at least
12 Mbit/s.
2. Balanced: This stream profile requires half of the bandwidth needed in the High quality
stream profile and seeks a balance between image quality and bandwidth usage. In this
case, the primary stream bandwidth may reach 4 Mbit/s, and each camera may require
more than 10 Mbit/s by also considering dual-channel traffic diversion and video
buffering. When only single-channel traffic diversion is required and video buffering is
not, the recommended bandwidth is at least 6 Mbit/s.
3. Low bandwidth: This stream profile requires half of the bandwidth needed in the
Balanced stream profile and ensures low bandwidth with acceptable image quality. In
this case, the primary stream bandwidth may reach 2 Mbit/s, and each camera may
require more than 5 Mbit/s by also considering dual-channel traffic diversion and video
not, the recommended bandwidth is at least 3 Mbit/s.
4. Mobile devices: This stream profile requires one quarter of the bandwidth needed in the
Low bandwidth stream profile and uses low resolution to be suited to mobile devices. In
this case, the primary stream bandwidth may reach 512 kbit/s, and each camera may
require more than 1.5 Mbit/s by also considering dual-channel traffic diversion and video

Huawei SDC
not, the recommended bandwidth is at least 768 kbit/s.
Figure 2-48 Stream profiles
2.5.14 Active Anti-Fogging

Multiple sensors are used to detect the temperature and humidity and calculate the
temperature differences. If fogging conditions are met, the fan or heating module is started to
protect the camera against fog.
2.5.15 One-Click Log Collection

The eDesk can collect device fault information in one-click mode. Fault information can be
collected by fault symptom and logs can be downloaded to ensure fast and accurate fault
locating.
2.5.16 Troubleshooting
When a fault occurs on the live network, fast troubleshooting is required. The troubleshooting
cases, fault diagnosis, and fault information collection functions help users quickly and
accurately locate network faults and improve onsite troubleshooting capabilities

Huawei SDC
Figure 2-49 Troubleshooting function
2.5.17 Invisible IR Illumination

In key areas near railway lines, such as the throat sections, flyovers across the railway line,
and tunnels, the night vision cameras must use invisible IR illumination for surveillance,
ensuring safe traveling of trains. Generally, night vision cameras use 850 nm IR lamps or 940
nm IR lamps. The 850 nm IR lamps will produce a visible red glow at night, while 940 nm IR
lamps provide completely covert illumination. Invisible IR illuminators avoid confusion
between the visible red glow and the traffic light. For train drivers suffering from driving
fatigue, that confusion may lead to emergency brakes, causing traffic safety risks. Invisible IR
lamps are recommended for night surveillance, which will produce good imaging effect while
ensuring the train traveling safety.
2.5.18 GPS
With the rapid development of intelligent applications and video big data, cameras nowadays
are no longer pure video collection devices. If users need to view, analyze, and perform
operations on traditional cameras in the same area or at specific sites, they need to identify
cameras at each site using On-Screen Display (OSD) and then group cameras manually.
However, for Huawei cameras, users can manually mark the cameras' location information or
locate the cameras using the built-in GPS module (with the BDS integrated). The backend
platform can visualize the camera installation locations on a map. Based on this function,
more applications, such as the multi-camera collaboration, can be developed.
A camera with the built-in GPS module also supports time calibration, which enhances time
precision for recordings and images captured offline.
2.5.19 Posture Detection

Posture changes of pole-mounted cameras due to the vibration caused by wind or passing
vehicles will lead to the change of the cameras' surveillance view. To avoid negative effects of

Huawei SDC
the posture changes, Huawei cameras are equipped with built-in G-sensors to monitor the
camera posture in real time. An alarm will be generated when the camera posture changes.
Working principle: When a camera is powered on, the system automatically obtains and
records the current posture of the camera and obtains the latest posture in real time. When
detecting that the posture change exceeds the preset threshold and the change persists, the
system generates an alarm.

Huawei SDC
Technical White Paper 3 Appendix A Acronyms and Abbreviations
3 Appendix A Acronyms and Abbreviations
Table 3-1 Acronyms and abbreviations
Acronym/Abbreviation Full Name
CPU Central Processing Unit
CU Client Unit
CPE Customer Premises Equipment
DSL Digital Subscriber Loop
DVS Digital Video Server

DVR Digital Video Recorder
EMC Electromagnetic Compatibility

eCN Evolution Core Network
GOP Group of Pictures

HME Hierarchical Motion Estimation
DHCP Dynamic Host Configuration Protocol

IP Internet Protocol
SDC Software Defined Camera
IVS Intelligent Video Surveillance
LAN Local Area Network

LTE Long Term Evolution
NVR Network Video Recorder

ONVIF Open Network Video Interface Forum
PC Personal Computer
PAL Phase Alternating Line

PTZ Pan/Tilt/Zoom

Huawei SDC
Technical White Paper 3 Appendix A Acronyms and Abbreviations
Acronym/Abbreviation Full Name
PU Peripheral Unit
RRU Remote Radio Unit

ROI Range of Interest
SD Secure Digital Memory Card
SFP Small Form-Factor Pluggable

SDK Software Development Kit
TDD Time Division Duplexing

Huawei SDC Technical White Paper - 290819

Uploaded by

Copyright:

Available Formats

Huawei SDC Technical White Paper - 290819

Uploaded by

Document Information

Original Title

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

Huawei SDC Technical White Paper - 290819

Uploaded by

Copyright:

Available Formats

Huawei SDC

Technical White Paper

Huawei Technologies Co., Ltd.

Trademarks and Permissions

Huawei Technologies Co., Ltd.

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. i

Change History ...............................................................................Error! Bookmark not defined.

2 Related Technologies ................................................................................................................... 4

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. ii

2.2.20 Software-Defined Camera ..................................................................................................................................... 28

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. iii

2.5.6 Multiple Power Supply Modes ................................................................................................................................ 51

3 Appendix A Acronyms and Abbreviations............................................................................ 59

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. iv

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 1

1.2 SDC Components

Image sensor (CCD or COMS)

Lens ICR module External ports

Figure 1-2 Video surveillance services

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 2

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 3

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 4

Figure 2-1 SEC technology implementation diagram

Orig inal data stream

Note: 1-1->1-2: no SEC processi ng 2-1->2-2: SEC processi ng added

2.1.3 Corridor Mode

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 5

Figure 2-2 Video image in 16:9 aspect ratio

Figure 2-3 Video image in 9:16 aspect ratio

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 6

2.1.4 High-Frame-Rate Encoding Technology

Figure 2-4 HD video at 30 fps

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 7

Figure 2-5 HD video at 60 fps

2.1.6 Five Streams

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 8

Figure 2-6 Setting video stream parameters

This feature applies to M12 and M22 series cameras.

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 9

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 10

2.2.2 Abnormal Event Detection

Figure 2-7 Tripwire crossing detection

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 11

Figure 2-8 Loitering detection

Figure 2-9 Intrusion detection

4. Abandoned object detection

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 12

Figure 2-10 Abandoned object detection

5. Removed object detection

Figure 2-11 Removed object detection

6. Area entry detection

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 13

Figure 2-12 Area entry detection

7. Area exit detection

Figure 2-13 Area exit detection

8. Fast movement detection

Issue V2.1 (2019-05-25) Copyright © Huawei Technologies Co., Ltd. 14

Figure 2-14 Fast movement detection

9. Crowd gathering detection