Feature Detection

We directly measure the evolution of the intergalactic Lyman-α effective optical depth, τ eff , over the redshift range 2 ≤ z ≤ 4.2 from a sample of 86 high-resolution, high-signal-to-noise quasar spectra obtained with the ESI and HIRES spectrographs on Keck, and with the MIKE spectrograph on Magellan. This represents an improvement over previous analyses of the Lyα forest from highresolution spectra in this redshift interval of a factor of two in the size of the data set alone. We pay particular attention to robust error estimation and extensively test for systematic effects. We find that our estimates of the quasar continuum levels in the Lyα forest obtained by spline fitting are systematically biased low, with the magnitude of the bias increasing with redshift, but that this bias can be accounted for using mock spectra. The mean fractional error ∆C/C true is < 1% at z = 2, 4% at z = 3, and 12% at z = 4. Previous measurements of τ eff at z 3 based on directly fitting the quasar continua in the Lyα forest, which have generally neglected this effect, are therefore likely biased low. We provide estimates of the level of absorption arising from metals in the Lyα forest based on both direct and statistical metal removal results in the literature, finding that this contribution is ≈ 6 -9% at z = 3 and decreases monotonically with redshift. The high precision of our measurement, attaining 3% in redshift bins of width ∆z =0.2 around z = 3, indicates significant departures from the best-fit power-law redshift evolution (τ eff = 0.0018(1 + z) 3.92 , when metals are left in), particularly near z = 3.2. The observed downward departure is statistically consistent with a similar feature detected in a precision statistical measurement using Sloan Digital Sky Survey spectra by Bernardi and coworkers using an independent approach.

Bathymetric mapping is essential for safe navigation, coastal infrastructure development, ecosystem monitoring, and understanding marine geological processes. As survey sensors become increasingly capable, the volume and resolution of bathymetric data have grown exponentially, exceeding the capacity of traditional manual analysis techniques. Conventional workflows, although proven, remain resource-intensive, subjective, and prone to inter-operator variability. Artificial intelligence (AI), including machine learning (ML) and deep learning (DL) models, offers transformative potential to automate feature detection, classification, and anomaly identification in bathymetric datasets. This research investigates the emerging role of AI in automating bathymetric data analysis, reviewing state-of-theart techniques such as convolutional neural networks, transfer learning, and unsupervised clustering. The paper highlights case studies demonstrating successful applications in shipwreck detection, habitat classification, and geomorphological segmentation, while also discussing critical challenges around data scarcity, generalization across diverse seafloor types, and the interpretability of AI models. Overall, the synthesis of current knowledge suggests that AI-supported bathymetric workflows can substantially improve operational efficiency, consistency, and accuracy for marine geospatial applications, while underscoring the need for further research in explainable and transferable frameworks.

Association rule mining (ARM) has been studied in the areas of content-based multimedia retrieval and semantic concept detection due to its high efficiency and accuracy. Two important processes in mining the association rules for classification are rule generation and rule selection. In this paper, a novel high-level feature detection framework using the ARM technique together with the correlations among the feature-value pairs is proposed. A new association rule mining (ARM) algorithm has been developed, where the N -featurevalue pairs are generated using a combined measure based on (1) the existence of the (N -1)-feature-value pairs (where N is larger than 1), (2) the correlation between different Nfeature-value pairs and the concept classes through Multiple Correspondence Analysis (MCA), and (3) the similarity representing the harmonic mean of the inter-similarity and intrasimilarity. The final association classification rules are selected by using the calculated harmonic mean of the similarity values. The proposed framework enables the automatic discovery and generation of the N -feature-value pair association rules from the 1-feature-value pairs for classification. Experimenting with 15 high-level features (concepts) and benchmark data sets from TRECVID, our proposed framework achieves promising performance and outperforms three other wellknown classifiers (Decision Trees, Support Vector Machine, and Neural Networks) which are commonly used for performance comparison in the TRECVID community.

Computer vision is much more than a technique to sense and recover environmental information from an UAV. It should play a main role regarding UAVs' functionality because of the big amount of information that can be extracted, its possible uses and applications, and its natural connection to human driven tasks, taking into account that vision is our main interface to world understanding. Our current research's focus lays on the development of techniques that allow UAVs to maneuver in spaces using visual information as their main input source. This task involves the creation of techniques that allow an UAV to maneuver towards features of interest whenever a GPS signal is not reliable or sufficient, e.g. when signal dropouts occur (which usually happens in urban areas, when flying through terrestrial urban canyons or when operating on remote planetary bodies), or when tracking or inspecting visual targets -including moving ones-without knowing their exact UMT coordinates. This paper also investigates visual servoing control techniques that use velocity and position of suitable image features to compute the references for flight control. This paper aims to give a global view of the main aspects related to the research field of computer vision for UAVs, clustered in four main active research lines: visual servoing and control, stereo-based visual navigation, image processing algorithms for detection and tracking, and visual SLAM. Finally, the results of applying these techniques in several applications are presented and discussed: this study will encompass power line inspection, mobile target tracking, stereo distance estimation, mapping and positioning. Keywords UAV • visual servoing • image processing • feature detection • tracking • SLAM

During 2004 the Cassini/RPWS (Radio and Plasma Wave Science) instrument recorded about 5400 SEDs (Saturn Electrostatic Discharges), which were organized in 4 storm systems and 95 episodes. A computer algorithm with different intensity thresholds was applied to extract the SEDs from the RPWS data, and a statistical analysis on the main characteristics of these SEDs is performed. Compared to the SEDs recorded by the Voyagers in the early 1980s, some characteristics like SED rate, intensity, signal duration, or power spectrum are similar, but there are also remarkable differences with regard to time occurrence and frequency range: The first appearance of SEDs (storm 0) was recorded by RPWS from a distance of more than 300 Saturn radii at the end of May 2004, followed by storm A in mid-July, storm B at the beginning of August, and the most prominent storm C throughout most of September. There were also significant intervals of time with no detectable SED activity, e.g., SEDs were practically absent from October 2004 until June 2005. No clear indication for SEDs below a frequency of 1.3 MHz could be found. We suggest that the SED storms A, B, C, and possibly also storm 0 originate from the same storm system residing at a latitude of 35 • South, which lasted for several months, waxed and waned in strength, and rotated with the Voyager radio period of Saturn. The SED source might be located in the updrafting water clouds beneath the visible cloud features detected in the Cassini images.

This paper proposes a content based image retrieval (CBIR) system using the local colour and texture features of selected image sub-blocks and global colour and shape features of the image. The image sub-blocks are roughly identified by segmenting the image into partitions of different configuration, finding the edge density in each partition using edge thresholding followed by morphological dilation. The colour and texture features of the identified regions are computed from the histograms of the quantized HSV colour space and Gray Level Co-occurrence Matrix (GLCM) respectively. The colour and texture feature vectors is computed for each region. The shape features are computed from the Edge Histogram Descriptor (EHD). A modified Integrated Region Matching (IRM) algorithm is used for finding the minimum distance between the sub-blocks of the query and target image. Experimental results show that the proposed method provides better retrieving result than retrieval using some of the existing methods.

Impact cratering is a ubiquitous process in our solar system affecting all planetary surfaces throughout geologic time. On Earth, there are currently 190 confirmed impact structures, which are distributed unevenly. The Fennoscandian Shield houses 17 % of them. The large amount of impact structures makes Fennoscandia one of the most densely cratered terrains on Earth. A dozen (12) impact structures have been discovered in Finland. The latest discovery, Lake Summanen is located in Central Finland, about 9 km southeast of city Saarijärvi. An impact generated structure was first hinted by airborne geophysical mapping by the Geological Survey of Finland in the early 2000`s (Lerssi et al., 2007) that revealed a circular ~2.6 km wide striking aeroelectromagnetic resistivity anomaly. Recent studies in 2017-2018 confirmed its impact origin based on the findings of shatter cone-bearing rocks and the identification of planar deformation features in quartz.

This paper presents the results of the SHREC&#39;19 track: Feature curve extraction on triangle meshes. Given a model, the challenge consists in automatically extracting a subset of the mesh vertices that jointly represent a feature curve. As an optional task, participants were requested to send also a similarity evaluation among the feature curves extracted. The various approaches presented by the participants are discussed, together with their results. The proposed methods highlight different points of view of the problem of feature curve extraction. It is interesting to see that it is possible to deal with this problem with good results, despite the different approaches.

Object localisation and identification is a crucial problem for advanced mobile service robots. We developed an object recognition system that localises and identifies objects using a colour-based visual attention control algorithm and a hierarchical neural network for object classification utilising hierarchical class grouping. The approach is evaluated in a test scenario where a robot is situated in front of a table. The robot has to identify and manipulate objects lying on this table. We evaluated the total object recognition performance and compared the effectiveness of different feature sets. The approach showed very encouraging results and meets real-time constraints.

In this paper, we discuss the possibility of using artificial neural networks (ANNs) as feature detectors in automatic target recognition (ATR). The goal is to discern a vehicle in an infrared image. We train ANNs to recognize the most easily recognizable parts of the vehicles, the wheels. The specific ANNs we use, shared weight ANNs, are especially adept at such an image recognition task due to their specialized architecture. The feature detection stage results in an image containing in each pixel the output of the ANN, indicating its confidence in the classification. We can then use a simple sequence of image processing algorithms on this image to find peaks and, by counting the number of these peaks, vehicles. This system is tested on sensitivity to scale differences and background clutter and is shown to perform quite well.

We propose a method for estimating depth from images captured with a real aperture camera by fusing defocus and stereo cues. The idea is to use stereobased constraints in conjunction with defocusing to obtain improved estimates of depth over those of stereo or defocus alone. The depth map as well as the original image of the scene are modeled as Markov random fields with a smoothness prior, and their estimates are obtained by minimizing a suitable energy function using simulated annealing. The main advantage of the proposed method, despite being computationally less efficient than the standard stereo or DFD method, is simultaneous recovery of depth as well as space-variant restoration of the original focused image of the scene.

Financial news carry information about economical figures and indicators. However, these texts are mostly unstructured and consequently hard to be processed in an automatic way. In this paper, we present a representation formalism that supports a linguistic composition for machine learning tasks. We show an innovative approach to structuring financial texts by extracting principal indicators. Considering announcements in the monetary policy domain, we distinguish between attributes and their values and argue that attributes are to be represented as an aggregated set of economic terms, keeping their values as corresponding conditional expressions. We close with a critical discussion and future perspectives.

Information is commonly reflected in news articles. However, texts are unstructured and thus demanding to analyze automatically. To identify and capture the facts in a news story we propose a novel approach, which utilizes natural language engineering. A combination of selected linguistic and statistical criteria enables the identification of grammatical units such as noun, verb, adjective, and adverb phrases. In literature, these entities are presumed to carry the meaning and the information expressed in English texts. In our study, we focus on determining multi-word features in articles related to the monetary policy conducted by the central bank in the USA, FED. The features are composed as attribute-value pairs, where the attributes represent grammatical units, which quantify the major event characteristics. The corresponding values are conditional expressions, which vary over time as facts evolve. The final set is aggregated over the corpus by the application of heuristic and syntax-based rules. Financial experts contributed to the project by providing expertise for the document interpretation.

A novel method for identification and extraction of multi-word features, defined as attribute-value pairs, from official press releases. For the specific purpose, we use linguistic and statistical criteria in conjunction with financial domain knowledge.

In Fall 2008 NASA selected a large international consortium to produce a comprehensive automated feature-recognition system for the Solar Dynamics Observatory (SDO). The SDO data that we consider are all of the Atmospheric Imaging Assembly (AIA) images plus surface magnetic-field images from the Helioseismic and Magnetic Imager (HMI). We produce robust, very efficient, professionally coded software modules that can keep up with the SDO data stream and detect, trace, and analyze numerous phenomena, in-The Solar Dynamics Observatory

Modern unsteady (multi-)field visualizations require an effective reduction of the data to be displayed. From a huge amount of information the most informative parts have to be extracted. Instead of the fuzzy application dependent notion of feature, a new approach based on information theoretic concepts is introduced in this paper to detect important regions. This is accomplished by extending the concept of local statistical complexity from finite state cellular automata to discretized (multi-)fields. Thus, informative parts of the data can be highlighted in an application-independent, purely mathematical sense. The new measure can be applied to unsteady multifields on regular grids in any application domain. The ability to detect and visualize important parts is demonstrated using diffusion, flow, and weather simulations.

Abstract: Many steganographic techniques were proposed for hiding secret message inside images, the simplest of them being the LSB data hiding. In this paper, we suggest a novel data hiding technique in an HTML Web page and also propose some simple techniques to extend the embedding technique to source codes written in any programming language (both case insensitive like HTML, Pascal and case sensitive languages like C, C++, Java). We basically try to exploit the case-redundancy in case-insensitive language, while we try ...

A moving visual figure may contain first-order signals defined by variation in mean luminance, as well as second-order signals defined by constant mean luminance and variation in luminance envelope, or higher-order signals that cannot be estimated by taking higher moments of the luminance distribution. Separating these properties of a moving figure to experimentally probe the visual subsystems that encode them is technically challenging and has resulted in debated mechanisms of visual object detection by flies. Our prior work took a white noise systems identification approach using a commercially available electronic display system to characterize the spatial variation in the temporal dynamics of two distinct subsystems for first-and higher-order components of visual figure tracking. The method relied on the use of single pixel displacements of two visual stimuli according to two binary maximum length shift register sequences (m-sequences) and cross-correlation of each m-sequence with time-varying flight steering measurements. The resultant spatio-temporal action fields represent temporal impulse responses parameterized by the azimuthal location of the visual figure, one STAF for first-order and another for higher-order components of compound stimuli. Here we review m-sequence and reverse correlation procedures, then describe our application in detail, provide Matlab code, validate the STAFs, and demonstrate the utility and robustness of STAFs by predicting the results of other published experimental procedures. This method has demonstrated how two relatively modest innovations on classical white noise analysis-the inclusion of space as a way to organize response kernels and the use of linear decoupling to measure the response to two channels of visual information simultaneously-could substantially improve our basic understanding of visual processing in the fly.

Tongue body features such as color is used in Traditional Chinese Medicine (TCM) practices to diagnose a patient&#39;s state of health. However, the diagnosis of one patient&#39;s health condition varies between practitioners. This gives rise to the need of a standard method such as using Support Vector Machine (SVM) to identify the tongue body color. Typically, SVM classifiers are implemented in software where the classification performance is very dependent on the architecture of the general-purpose CPU. Since classification of tongue images is a recurring event, the design of a hardware accelerator is proposed in this project. The purpose of designing a hardware accelerator is to boost the classifier performance, execution time and latency so that it meets real-time constraints. Architectural optimization methods, such as loop unrolling, memory array partitioning and pipelining of the SVM classification algorithm are utilized and the final hardware architecture is synthesized to ...

We introduce PixelFlex2, our newest scalable wall-sized, multi-projector display system. For it, we had to solve most of the difficult problems left open by its predecessor, PixelFlex, a proof-of-concept demonstration driven by a large, multi-headed SGI graphics system. PixelFlex2 retains the achievements of PixelFlex (high-performance through single-pass rendering, single-pixel accuracy for geometric blending with only casual placement of projectors), while adding a) higher performance and scalability with a Linux PC-cluster, b) ...

The purpose of this project is the creation of a controller for an autonomous robot. To achieve this, the accelerometer of the STM32F4 - Dsicovery board will be used which will be coupled to the robot. This project it is an extension of pre vious projects referenced below which worked with the same robot implementing the following application: - Design and i mplementation of a line tracking system for the robot. - Design and implementation of a WiFi communication system for the robot. The extensio n will be the design and implementation of a PD controller for the line tracking robot. The innovation will be the use of the acceleration data obtained directly from the accelerometer MEMS of the discovery board in order to design the derivative controlle r. First step will be the theoretical design of the controller using the Routh - Hurwitz theorem and the following simulations using Matlab an d Simulink, to prove the response of the robot that will be expected in the experimental tests...

The method of surrogate data, classically used for nonlinearity tests, amounts to the use of some constrained noise providing reference for statistical test. It is revisited here as a method for stationarization and this feature is put forward in the context of nonstationarity testing. The stationarization property of surrogates is first explored in a time-frequency perspective and used for devising a test of stationarity relative to an observation time. Then, more general forms of surrogates are developed, directly in time-frequency or mixed domains of representation (ambiguity and time-lag domains included) and it is shown how they allow for other tests of nonstationary features: detection of the existence of a transient in some noise; assessment of nonstationary cross-correlations.

Nowadays, video recordings are widely used and ease to spread the knowledge among the students. Due to the rapid development of recording technologies and video based learning, the large number of videos is published in the Internet. The main challenge is to retrieve the appropriate video based on the user requirement. To achieve this objective, the newly designed intent aware optimization is proposed in this paper using grey wolf optimizer. The extraction of key frame is the initial step in the proposed system. Then, based on the input query, the keywords from the key frame are recognized by the optical character recognition and LVP. After the features are extracted, the PENN classifier is utilized to retrieve the relevant videos for the text or video query. Subsequently, the user selects one video which is used for the matching purpose based on the optimization. Thus, the intent aware optimization is newly designed using the grey wolf optimizer. The grey wolf optimizer is applied ...

This work presents the application of a basic unsupervised classification algorithm for the segmentation of indoor passive Terahertz images. The 30,000 pixel broadband images of a person with concealed weapons under clothing are taken at a range of 0.8-2 m over a frequency range of 0.1-1.2 THz using single-pixel row-based raster scanning. The spiralantenna coupled 36x1x0.02 µm Nb bridge cryogenic micro-bolometers are developed at NIST-Optoelectronics Division. The antenna is evaporated on a 250 µm thick Si substrate with a 4 mm diameter hyper-hemispherical Si lens. The NETD of the microbolometer is 125 mK at an integration time of 30 ms. The background temperature calibration is performed with a known 25 pixel source above 330 K, and a measured background fluctuation of 200-500 mK. Several weapons were concealed under different fabrics: cotton, polyester, windblocker jacket and thermal sweater. Measured temperature contrasts ranged from 0.5-1 K for wrinkles in clothing to 5 K for a zipper and 8 K for the concealed weapon. In order to automate feature detection in the images, some image processing and pattern recognition techniques have been applied and the results are presented here. We show that even simple algorithms, that can potentially be performed in real time, are capable of differentiating between a metal and a dielectric object concealed under clothing. Additionally, we show that pre-processing can reveal low temperature contrast features, such as folds in clothing.

Emerging mobile applications, such as augmented reality, demand robust feature detection at high frame rates. We present an implementation of the popular Scale-Invariant Feature Transform (SIFT) feature detection algorithm that incorporates the powerful graphics processing unit (GPU) in mobile devices. Where the usual GPU methods are inefficient on mobile hardware, we propose a heterogeneous dataflow scheme. By methodically partitioning the computation, compressing the data for memory transfers, and taking into account the unique challenges that arise out of the mobile GPU, we are able to achieve a speedup of 4-7x over an optimized CPU version, and a 6.4x speedup over a published GPU implementation. Additionally, we reduce energy consumption by 87 percent per image. We achieve near-realtime detection without compromising the original algorithm.

Face is a complex multidimensional structure and needs good computing techniques for recognition. Our approach consists of two feature extraction algorithms that are Gabor wavelet and local binary pattern for the purpose of color face recognition. These methods provide excellent recognition rates for face images taken under severe variation in pose, illumination as well as for small resolution face images. Face recognition is done by Principal Component Analysis. This work shows the performance of color face recognition with YCbCr color space using GW and LBP. The experiments are performed on FEI color database. Result includes recognition rate (in percent) for different method such as YCbCr and Gray using GW and LBP for different pixels resolution such as 54x36, 81x54, and 108x72.

We investigate fast optimum route planning in large, real-world road networks for car navigation systems. We show how graph partitioning can be used to increase the speed of planning optimum routes. Creating a graph partition with future route planning in mind leads to a non-standard graph partitioning problem. In particular, the quality of a partition, indicated by the objective value, is assumed to represent the execution time of the route planning process. We present an efficient approximation algorithm for creating graph partitions suited for fast optimum route planning. We study the relation between the objective value and the number of edges evaluated by the route planning algorithm, which is an objective measure of the route planning speed. Experiments show that the best partition according to the objective value does not lead to the fastest route planning process. We present a new objective value and show that better partitions result in faster route planning for our new objective value.

Recognition of any individual is a task to identify people. Human identification using Gait is method to identify an individual by the way he walk or manner of moving on foot. Gait recognition is a type of biometric recognition and related to the behavioural characteristics of biometric recognition. Gait recognition is one kind of biometric technology that can be used to monitor people without their cooperation. Controlled environments such as banks, military installations and even airports need to be able to quickly detect threats and provide differing levels of access to different user groups. Gait shows a particular way or manner of moving on foot and gait recognition is the process of identifying an individual by the manner in which they walk. Gait is less unobtrusive biometric, which offers the possibility to identify people at a distance, without any interaction or co-operation from the subject; this is the property which makes it so attractive. In this thesis, firstly binary ...

In this paper we present a new method, called TracMass, for analyzing data obtained using hyphenated chromatography-mass spectrometry (XC/MS). The method uses a Kalman filter to extract pure, noise-free ion chromatograms by exploiting the latent second order structure in the XC/MS data. TracMass differs from current state-of-the-art methodologies, which extract chromatograms by binning along the m/z axis and further processes the data in various ways, e.g. by baseline correction, component detection algorithm, peak detection, and curve resolution to extract molecular features. The proposed method was validated by analyzing two plasma datasets: one derived from 99 quality control samples where TracMass extracted 8880 Pure Ion Chromatograms (PICs) present in ≥90 of the samples. The second dataset was spiked with two different internal standard mixtures to test differential expression analysis. Here TracMass found 20 000 PICs present in 10 samples, all differentially expressed analytes, and also a previously unreported discriminating metabolite. Finding as many PICs as possible is in this context essential to ensure that even small differentiating features are found (if they exist). The resulting data representation from TracMass (PICs) can be used directly for statistical analysis, and the method is fast (approximately 5 min/sample), with few adjustable parameters.

Background: Mass Spectrometry coupled to Liquid Chromatography (LC-MS) is commonly used to analyze the protein content of biological samples in large scale studies. The data resulting from an LC-MS experiment is huge, highly complex and noisy. Accordingly, it has sparked new developments in Bioinformatics, especially in the fields of algorithm development, statistics and software engineering. In a quantitative label-free mass spectrometry experiment, crucial steps are the detection of peptide features in the mass spectra and the alignment of samples by correcting for shifts in retention time. At the moment, it is difficult to compare the plethora of algorithms for these tasks. So far, curated benchmark data exists only for peptide identification algorithms but no data that represents a ground truth for the evaluation of feature detection, alignment and filtering algorithms. We present LC-MSsim, a simulation software for LC-ESI-MS experiments. It simulates ESI spectra on the MS level. It reads a list of proteins from a FASTA file and digests the protein mixture using a user-defined enzyme. The software creates an LC-MS data set using a predictor for the retention time of the peptides and a model for peak shapes and elution profiles of the mass spectral peaks. Our software also offers the possibility to add contaminants, to change the background noise level and includes a model for the detectability of peptides in mass spectra. After the simulation, LC-MSsim writes the simulated data to mzData, a public XML format. The software also stores the positions (monoisotopic m/z and retention time) and ion counts of the simulated ions in separate files. LC-MSsim generates simulated LC-MS data sets and incorporates models for peak shapes and contaminations. Algorithm developers can match the results of feature detection and alignment algorithms against the simulated ion lists and meaningful error rates can be computed. We anticipate that LC-MSsim will be useful to the wider community to perform benchmark studies and comparisons between computational tools.

Recent work in the chemistry of molecular clouds in the early universe has found reactions that may create ancient water molecules that predate our Sun. For these to take place, particular conditions of molecular density and temperature must exist. Water molecules will be destroyed by high temperatures, or when densities are insufficient to shield them from UV rays. If the environment is too cold, the reactions will not take place. Water molecules will only be created and persist in areas of the universe in which these properties are suitable, and will do so at different rates depending on local conditions. In this work we combine a high resolution hydrodynamics simulation of the early universe closely coupled with this hydroxyl and water-producing chemistry model to determine how water molecules would be created and distributed in space and time in the early universe. This work is enabled by the computational power of todays supercomputers and simulation technology. The complexity ...

In this paper we describe our experiments in the automatic and interactive search tasks of TRECVID 2008. We submitted six runs, five of them are automatic and one is interactive. The automatic runs include, a text baseline, two runs based on visual features, and two runs that combine high level features and visual features. For our interactive search submission we developed a search interface based on both textual and low level features, called the Group Interface.

We present the main observational features expected for Gamma-Ray Bursts (GRBs) that occur inside pulsar wind bubbles (PWBs). This is the most natural outcome of the supranova model, where initially a supernova (SN) explosion takes place, leaving behind a supra-massive neutron star, which loses its rotational energy over a time t sd and collapses to a black hole, triggering a GRB. We find that the time delay, t sd , between the SN and GRB events is the most important parameter that determines the behavior of the system. We consider the afterglow, prompt GRB and PWB emission. Constraints on the model are derived for a spherical PWB, from current afterglow observations and the lack of direct detection of the PWB emission. We find that a simple spherical model cannot account for the X-ray features detected in some afterglows, together with the typical afterglow emission that was observed for the same GRBs. The discrepancies with observations may be reconciled by resorting to a non-spherical geometry, where the PWB is elongated along the polar axis. Finally, we predict that the inverse Compton upscattering of PWB photons by the relativistic electrons of the afterglow (external Compton) should lead to high energy emission during the early afterglow that may explain the GeV photons detected by EGRET for a few GRBs, and should be detectable by future missions such as GLAST.

For robots navigating using only a camera, illumination changes in indoor environments can cause re-localization failures during autonomous navigation. In this paper, we present a multisession visual SLAM approach to create a map made of multiple variations of the same locations in different illumination conditions. The multi-session map can then be used at any hour of the day for improved re-localization capability. The approach presented is independent of the visual features used, and this is demonstrated by comparing re-localization performance between multi-session maps created using the RTAB-Map library with SURF, SIFT, BRIEF, BRISK, KAZE, DAISY and SuperPoint visual features. The approach is tested on six mapping and six localization sessions recorded at 30 minute intervals during sunset using a Google Tango phone in a real apartment.

Persoalan deteksi multi fitur sketsa wajah adalah objek non rigid yang menggunakan modalitas berbeda. Persoalan tersebut kurang tepat diselesaikan menggunakan metode yang berbasis appearance, karena jarak antara fitur citra sketsa dan foto dari orang yang sama, lebih besar dibandingkan dengan jarak fitur foto dari dua orang yang berbeda. Untuk mendeteksi objek yang non-rigid dengan modalitas yang berbeda perlu disusun suatu model deformable yang dapat bekerja pada modalitas berbeda. Berdasarkan kondisi ...

Modality reduction by using the Eigentransform method can not efficiently work, when number of training sets larger than image dimension. While modality reduction by using the first derivative negative followed by feature extraction using Two Dimensional Discrete Cosine Transform has limitation, which is feature extraction achieved of face sketch feature is included non-dominant features. We propose to select the image region that contains the dominant features. For each region that contains dominant features will be extracted one frequency by using Two Dimensional-Discrete Cosine Transform. To reduce modality between photographs as training set and face sketches as testing set, we propose to bring the training and testing set toward new dimension by using the first derivative followed by negative process. In order to improve final result on the new dimension, it is necessary to add the testing set pixels by using the difference of photograph average values as training sets and the corresponding face sketches average as testing sets. We employed 100 face sketches as testing and 100 photographs as training set. Experimental results show that maximum recognition is 93%.

Python programming language has experienced a great progress and growing use in the scientific community in the last years as well as a direct impact on solar physics. Python is a very mature language and almost any fundamental feature you might want to do is already implemented in a library or module. SunPy is a common effort of, using the advantages of Python, developing tools to be applied for processing and analysis of solar data. In this work we present a particular development, based on Python, for the analysis of proper motions in time series of images through the local correlation tracking algorithm. A graphic user interface allows to select different parameters for the computations, visualization and analysis of flow fields. Python is an easy to learn, powerful programming language. It has efficient high-level data structures and a simple but effective approach to object-oriented programming. Python is an ideal language for scripting and rapid application development in man...

This paper considers an application of scale-invariant feature detection using scale-space analysis suitable for use with wide field of view cameras. Rather than obtain scalespace images via convolution with the Gaussian function on the image plane, we map the image to the sphere and obtain scale-space images as the solution to the heat (diffusion) equation on the sphere which is implemented in the frequency domain using spherical harmonics. The percentage correlation of scale-invariant features that may be matched between any two wide-angle images subject to change in camera pose is then compared using each of these methods. We also present a means by which the required sampling bandwidth may be determined and propose a suitable anti-aliasing filter which may be used when this bandwidth exceeds the maximum permissible due to computational requirements. The results show improved performance using scale-space images obtained as the solution of the diffusion equation on the sphere, with additional improvements observed using the anti-aliasing filter.

Background/Aims: The natural outcome of ultrasound-detected macronodules in cirrhosis is still poorly understood. In this study we assessed the incidence and predictors of malignant transformation in a prospective study of 90 consecutive ultrasound-detected macronodules in cirrhosis. Methods: Macronodules classification was based on recently proposed histological criteria. Extranodular large (LCC) and small cell changes were also evaluated. The follow-up included ultrasound and serum alfa-fetoprotein determination every 3 months. Independent predictors of hepatocellular carcinoma were evaluated by Cox proportional hazards regression analysis. Results: During a mean follow-up of 33 months, 28 (31%) nodules transformed into hepatocellular carcinoma. The incidence of hepatocellular carcinoma per 100 person-years of follow-up was 11.3%, with a malignant transformation rate of 3.5, 15.5, 31 and 48.5% at 1, 2, 3, and 5 years respectively. High-grade dysplastic nodules (HGDN) (hazard risk 5 2.4; CI 95% 5 1.1-5.0) and LCC (hazard risk 5 3.1; CI 95% 5 1.2 -7.8) were independent predictors of malignant transformation. Eight additional hepatocellular carcinomas developed outside the original lesions raising the overall malignant transformation rate to 40% while 15 macronodules (17%) became undetectable at ultrasound (US). Conclusions: Macronodules characterize a cirrhotic subpopulation with high risk of hepatocellular carcinoma. HGDN and LCC are strong predictors of malignant transformation; subjects with simultaneous presence of both these two conditions are at highest risk of cancer development. The management of cirrhotics with macronodules should be based on morphologic features detected on liver microsamples.

sis) owns certain copyright or related rights in it (the "Copyright") and s/he has given The University of Manchester certain rights to use such Copyright, including for administrative purposes. (ii) Copies of this thesis, either in full or in extracts and whether in hand or electronic copy, may be made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part of any such copies made. (iii) The ownership of certain Copyright, patents, designs, trade marks and other intellectual property (the "Intellectual Property") and any reproductions of copyright works in the thesis, for example graphs and tables ("Reproductions"), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproductions.