No description available.

No description available.

International Telemetering Conference Proceedings / October 20-23, 2003 / Riviera Hotel and Convention Center, Las Vegas, Nevada / We develop novel methods for compressing volumetric imagery that has been generated by single platform (mobile) range sensors. We exploit the correlation structure inherent in multiple views in order to improve compression efficiency. We show that, for lossy compression, three-dimensional volumes compress more efficiently than two-dimensional (2D) images. In fact, our error metric suggests that accumulating more than 9 range images in one volume before compression yields up to a 99% improvement in compression performance over 2D compression.

No description available.

L’utilisation de lasers pulsés ouvre la voie à de nouvelles fonctionnalités et à une compacité accrue des systèmes lidars pour la mesure de vibration à distance. Or des bruits de phase et d’amplitude affectent le signal lidar, diminuant particulièrement les performances du régime impulsionnel à multiplets, concept par ailleurs prometteur pour la mesure à longue portée.Ces travaux portent d’abord sur la caractérisation expérimentale de ces bruits afin de les modéliser, en particulier l’effet de la turbulence atmosphérique. Puis nous cherchons à optimiser les formes d’ondes et le traitement du signal en fonction de la vibration et de la statistique de bruit. Nous proposons une méthode originale basée sur un estimateur du maximum de vraisemblance de la fréquence Doppler, associé à une extraction à partir de la représentation temps-fréquence du signal. L’apport de cette approche est constaté par la simulation et l’expérience, en comparant les performances de plusieurs régimes d’émission. / The use of pulsed lasers could lead to new functionnalities and increased compacity of lidar systems for remote vibration sensing. However, specific amplitude and phase noises affect the lidar signal, and particularly decrease the performance of a polypulse based emission regime, thought to be promising for very long range measurements.This work first deals with the experimental characterization of these noise sources in order to properly model them, with a focus on atmospheric turbulence. We then seek to optimize the employed waveform and signal processing in regard of the vibration and noise conditions. An original method is proposed that involves maximum likelihood based estimation of the vibration-induced Doppler shift, and its extraction from a time-frequency representation of the signal. The benefits of this approach are shown in simulation and experimentation, by comparing the performance of various emission modes.

Pattern theory, a mathematical framework for representing knowledge of complex patterns developed by applied mathematician Ulf Grenander, has been shown to have potential uses in automatic target recognition (ATR). Prior research performed in the mid-1990s at Washington University in St. Louis resulted in ATR algorithms based on concepts in pattern theory for forward-looking infrared (FLIR) and laser radar (LADAR) imagery, but additional work was needed to create algorithms that could be implemented in real ATR systems. This was due to performance barriers and a lack of calibration between target models and real data. This work addresses some of these issues by exploring techniques that can be used to create practical pattern-theoretic ATR algorithms. This dissertation starts by reviewing the previous pattern-theoretic ATR research described above and discussing new results involving the unification of two previously separate outcomes of that research: multi-target detection/recognition and thermal state estimation in FLIR imagery. To improve the overall utility of pattern-theoretic ATR, the following areas are re-examined: 1) generalized diffusion processes to update target pose estimates and 2) the calibration of thermal models with FLIR target data. The final section of this dissertation analyzes the fundamental accuracy limits of target pose estimation under different sensor conditions, independent of the target detection/recognition algorithm employed. The Cramér-Rao lower bound (CRLB) is used to determine these accuracy limits.

This thesis concerns detection and recognition of ground object using data from laser radar systems. Typical ground objects are vehicles and land mines. For these objects, the orientation and articulation are unknown. The objects are placed in natural or urban areas where the background is unstructured and complex. The performance of laser radar systems is analyzed, to achieve models of the uncertainties in laser radar data. A ground object recognition method is presented. It handles general, noisy 3D point cloud data. The approach is based on the fact that man-made objects on a large scale can be considered be of rectangular shape or can be decomposed to a set of rectangles. Several approaches to rectangle fitting are presented and evaluated in Monte Carlo simulations. There are error-in-variables present and thus, geometric fitting is used. The objects can have parts that are subject to articulation. A modular least squares method with outlier rejection, that can handle articulated objects, is proposed. This method falls within the iterative closest point framework. Recognition when several similar models are available is discussed. The recognition method is applied in a query-based multi-sensor system. The system covers the process from sensor data to the user interface, i.e., from low level image processing to high level situation analysis. In object detection and recognition based on laser radar data, the range value’s accuracy is important. A general direct-detection laser radar system applicable for hard-target measurements is modeled. Three time-of-flight estimation algorithms are analyzed; peak detection, constant fraction detection, and matched filter. The statistical distribution of uncertainties in time-of-flight range estimations is determined. The detection performance for various shape conditions and signal-tonoise ratios are analyzed. Those results are used to model the properties of the range estimation error. The detector’s performances are compared with the Cramér-Rao lower bound. The performance of a tool for synthetic generation of scanning laser radar data is evaluated. In the measurement system model, it is possible to add several design parameters, which makes it possible to test an estimation scheme under different types of system design. A parametric method, based on measurement error regression, that estimates an object’s size and orientation is described. Validations of both the measurement system model and the measurement error model, with respect to the Cramér-Rao lower bound, are presented.

<p>With numerous applications in both military and civilian life, the demand for accurate 3D models of real world environments increases rapidly. Using an airborne laser scanner for the raw data acquisition and robust methods for data processing, the researchers at the Swedish Defence Research Agency (FOI) in Linköping hope to fully automate the modeling process.</p><p>The work of this thesis has mainly been focused on three areas: ground estimation, image segmentation and classification. Procedures have in each of these areas been developed, leading to a new algorithm for ground estimation, a number of segmentation methods as well as a full comparison of various decision values for an object based classification. The ground estimation algorithm developed has yielded good results compared to the method based on active contours previously elaborated at FOI. The computational effort needed by the new method has been greatly reduced compared to the former, as performance, particularly in urban areas, has been improved. The segmentation methods introduced have shown promising results in separating different types of objects. A new set of decision values and descriptors for the object based classifier has been suggested, which, according to tests, prove to be more efficient than the set p reviously used.</p> / <p>Med många tillämpningar både inom det civila och militära, ökar efterfrågan på noggranna och korrekta omvärldesmodeller snabbt. Forskare på FOI, Totalförsvarets Forskningsinstitut, arbetar med att fullt ut kunna automatisera den process som genererar dessa tredimensionella modeller av verkliga miljöer. En luftburen laserradar används för datainsamlingen och robusta metoder är under ständig utveckling för den efterföljande databehandlingen.</p><p>Arbetet som presenteras i denna rapport kan delas in i tre huvudområden: skattning av markyta, segmentering av data samt klassificering. Metoder inom varje område har utvecklats vilket lett fram till en ny algoritm för markestimering, en rad metoder för segmentering samt en noggrann jämförelse av olika beslutsvärden för en objektbaserad klassificering. Markskattningsalgoritmen har visat sig vara effektiv i jämförelse med en metod baserad på aktiva konturer som sedan tidigare utvecklats på FOI. Beräkningsbördan för den nya metoden är endast en bråkdel av den förra, samtidigt som prestandan, särskilt i urbana miljöer, har kunnat förbättras.</p><p>De segmenteringsmetoder som introducerats har visat på lovande resultat vad gäller möjligheten att särskilja olika typer av objekt. Slutligen har en ny uppsättning deskriptorer och beslutsvärden till den objektbaserade klassificeraren föreslagits. Den har enligt de tester som presenteras i rapporten visats sig vara mer effektiv än den uppsättning som använts fram till idag.</p>

<p>The main goal of this thesis is to show the strength of combining datasets from two different types of sensors to find anomalies in their data. The sensors used in this thesis are a hyperspectral camera and a scanning 3-D laser.</p><p>The report can be divided into two main parts. The first part discusses the properties of one of the datasets and how these are used to isolate anomalies. An issue to deal with here is not only what properties to look at, but how to make the process automatic. The information retained from the first dataset is then used to make intelligent choices in the second dataset. Again, one of the challenges is to make this process automatic and accurate. The second part of the project consists of presenting the results in a way that gives the most information to the user. This is done with a graphical user interface that allows the user to manipulate the way the result is presented.</p><p>The conclusion of this project is that the information from the combined sensor datasets gives better results than the sum of the information from the individual datasets. The key of success is to play to the strengths of the datasets in question. An important block of the work in this thesis, the calibration of the two sensors, was completed by Kevin Chan as his thesis work in Electrical Engineering at the University of Lund. His contribution gave access to calibrated data that supported the results presented in this report.</p>

The main goal of this thesis is to show the strength of combining datasets from two different types of sensors to find anomalies in their data. The sensors used in this thesis are a hyperspectral camera and a scanning 3-D laser. The report can be divided into two main parts. The first part discusses the properties of one of the datasets and how these are used to isolate anomalies. An issue to deal with here is not only what properties to look at, but how to make the process automatic. The information retained from the first dataset is then used to make intelligent choices in the second dataset. Again, one of the challenges is to make this process automatic and accurate. The second part of the project consists of presenting the results in a way that gives the most information to the user. This is done with a graphical user interface that allows the user to manipulate the way the result is presented. The conclusion of this project is that the information from the combined sensor datasets gives better results than the sum of the information from the individual datasets. The key of success is to play to the strengths of the datasets in question. An important block of the work in this thesis, the calibration of the two sensors, was completed by Kevin Chan as his thesis work in Electrical Engineering at the University of Lund. His contribution gave access to calibrated data that supported the results presented in this report.