The move from Standard Definition (SD) to High Definition (HD) represents a six times increases in data, which needs to be processed. With expanding resolutions and evolving compression, there is a need for high performance with flexible architectures to allow for quick upgrade ability. The technology advances in image display resolutions, advanced compression techniques, and video intelligence. Software implementation of these systems can attain accuracy with tradeoffs among processing performance (to achieve specified frame rates, working on large image data sets), power and cost constraints. There is a need for new architectures to be in pace with the fast innovations in video and imaging. It contains dedicated hardware implementation of the pixel and frame rate processes on Field Programmable Gate Array (FPGA) to achieve the real-time performance. The following outlines the contributions of the dissertation. (1) We develop a target detection system by applying a novel running average mean threshold (RAMT) approach to globalize the threshold required for background subtraction. This approach adapts the threshold automatically to different environments (indoor and outdoor) and different targets (humans and vehicles). For low power consumption and better performance, we design the complete system on FPGA. (2) We introduce a safe distance factor and develop an algorithm for occlusion occurrence detection during target tracking. A novel mean-threshold is calculated by motion-position analysis. (3) A new strategy for gesture recognition is developed using Combinational Neural Networks (CNN) based on a tree structure. Analysis of the method is done on American Sign Language (ASL) gestures. We introduce novel point of interests approach to reduce the feature vector size and gradient threshold approach for accurate classification. (4) We design a gesture recognition system using a hardware/ software co-simulation neural network for high speed and low memory storage requirements provided by the FPGA. We develop an innovative maximum distant algorithm which uses only 0.39% of the image as the feature vector to train and test the system design. Database set gestures involved in different applications may vary. Therefore, it is highly essential to keep the feature vector as low as possible while maintaining the same accuracy and performance

Nous proposons dans cette thèse une méthode robuste de détection d’objets mobiles depuis une caméra en mouvement montée sur un vecteur aérien de type drone ou hélicoptère. Nos contraintes industrielles sont particulièrement fortes : robustesse aux grands mouvements de la caméra, robustesse au flou de focus ou de bougé, et précision dans la détection et segmentation des objets mobiles. De même, notre solution doit être optimisée afin de ne pas être trop consommatrice en termes de puissance de calcul. Notre solution consiste en la compensation du mouvement global, résultant du mouvement de la caméra, puis en l’analyse du mouvement résiduel existant entre les images pour détecter et segmenter les cibles mobiles. Ce domaine a été particulièrement exploré dans la littérature, ce qui se traduit par une richesse des méthodes proposées fondamentalement différentes. Après en avoir étudié un certain nombre, nous nous sommes aperçus qu’elles avaient toutes un domaine d’applications restreint, malheureusement incompatible avec nos préoccupations industrielles. Pour pallier à ce problème, nous proposons une méthodologie consistant à analyser les résultats des méthodes de l’état de l’art de manière à en comprendre les avantages et inconvénients de chacune. Puis, des hybridations de ces méthodes sont alors mis en place. Ainsi, nous proposons trois étapes successives : la compensation du mouvement entre deux images successives, l’élaboration d’un arrière plan de la scène afin de pouvoir segmenter de manière correcte les objets mobiles dans l’image et le filtrage de ces détections par confrontation entre le mouvement estimé lors de la première étape et le mouvement résiduel estimé par un algorithme local. La première étape consiste en l’estimation du mouvement global entre deux images à l’aide d’une méthode hybride composée d’un algorithme de minimisation ESM et d’une méthode de mise en correspondance de points d’intérêt Harris. L’approche pyramidale proposée permet d’optimiser les temps de calcul et les estimateursrobustes (M-Estimateur pour l’ESM et RANSAC pour les points d’intérêt) permettent de répondre aux contraintes industrielles. La deuxième étape établit un arrière plan de la scène à l’aide d’une méthode couplant les résultats d’une différence d’images successives (après compensation) et d’une segmentation en régions. Cette méthode réalise une fusion entre les informations statiques et dynamiques de l’image. Cet arrière plan est ensuite comparé avec l’image courante afin de détecter les objets mobiles. Enfin, la dernière étape confronte les résultats de l’estimation de mouvement global avec le mouvement résiduel estimé par un flux optique local Lucas-Kanade afin de valider les détections obtenues lors de la seconde étape. Les expériences réalisées dans ce mémoire sur de nombreuses séquences de tests (simulées ou réelles) permettent de valider la solution retenue. Nous montrons également diverses applications possibles de notre méthode proposée. / We propose a robust method about moving target detection from a moving UAV-mounted or helicopter-mounted camera. The industrial solution has to be robust to large motion of the camera, focus and motion blur in the images, and need to be accurate in terms of the moving target detection and segmentation. It does not have to need a long computation time. The proposed solution to detect the moving targets consists in the global camera motion compensation, and the residual motion analysis, that exists between the successive images. This research domain has been widely explored in the literature, implying lots of different proposed methods. The study of these methods show us that they all have a different and limited application scope, incompatible with our industrial constraints. To deal with this problem, we propose a methodology consisting in the analysis of the state-of-the-art method results, to extract their strengths and weaknesses. Then we propose to hybrid them. Therefore, we propose three successive steps : the inter-frame motion compensation, thecreation of a background in order to correctly detect the moving targets in the image and then the filtering of these detections by a comparison between the estimated global motion of the first step and the residual motion estimated by a local algorithm. The first step consists in the estimation of the global motion between two successive images thanks to a hybrid method composed of a minimization algorithm (ESM) and a feature-based method (Harris matching). The pyramidal implementation allows to optimize the computation time and the robust estimators (M-Estimator for the ESM algorithm and RANSAC for the Harris matching) allow to deal with the industrial constraints. The second step createsa background image using a method coupling the results of an inter-frame difference (after the global motion compensation) and a region segmentation. This method merges the static and dynamic information existing in the images. This background is then compared with the current image to detect the moving targets. Finally, the last step compares the results of the global motion estimation with the residual motion estimated by a Lucas-Kanade optical flow in order to validate the obtained detections of the second step. This solution has been validated after an evaluation on a large number of simulated and real sequences of images. Additionally, we propose some possible applications of theproposed method.

La détection et le suivi de cibles sont des missions fréquentes pour la robotique mobile, que le contexte soit civil, industriel ou militaire. Ces applications constituent un domaine de choix pour la planification multirobot, et sont abordées par de multiples communautés selon différents points de vue. Nous proposons dans un premier temps une taxonomie commune qui permetde regrouper et de comparer les différentes approches de ces problèmes, afin de mieux les analyser et de mettre en évidence leurs lacunes respectives. En particulier, on note la faible représentativité des modèles exploités, peu expressifs : la plupart des algorithmes évoluent dans un monde en deux dimensions où les observations et le déplacement sont conditionnés par lesmêmes obstacles. Ces modèles éloignés de la réalité nous semblent trop restrictifs pour pleinement exploiter la synergie des équipes multirobot hétérogènes : nous proposons une organisation des différents modèles nécessaires, en explicitant une séparation claire entre modèles et algorithmes de planification. Cette organisation est concrétisée par une librairie qui structure lesmodèles disponibles et définit les requêtes nécessaires aux algorithmes de planification. Dans un second temps, nous proposons un ensemble d’algorithmes utilisant les modèles définis précédemment pour planifier des missions de patrouille de zones et de poursuite de cibles. Ces algorithmes s’appuient sur un formalisme mathématique rigoureux afin d’étudier l’impact des modèlessur les performances. Nous analysons notamment l’impact sur la complexité – c’est-à-dire en quoi des modèles plus élaborés impactent la complexité de résolution – et sur la qualité des solutions résultantes, indépendamment des modèles, selon des métriques usuelles. D’une manière plus générale, les modèles sont un lien essentiel entre l’Intelligence Artificielle et la Robotique : leur enrichissement et leur étude approfondie permettent d’exhiber des comportements plus efficaces pour la réussite des missions allouées aux robots. Cette thèse contribue à démontrer l’importance des modèles pour la planification et la conduite de mission multirobots. / Detecting, localizing or following targets is at the core of numerous robotic applications in industrial, civilian and military application contexts. Much work has been devoted in various research communities to planning for such problems, each community with a different standpoint. Our thesis first provides a unifying taxonomy to go beyond the frontiers of specific communities and specific problems, and to enlarge the scope of prior surveys. We review various work related to each class of problems identified in the taxonomy, highlighting the different approaches, models and results. This analysis specifically points out the lack of representativityof the exploited models, which are in vast majority only 2D single-layer models where motion and sensing are mixed up. We consider those unrealistic models as too restrictive to handle the full synergistic potential of an heterogeneous team of cooperative robots. In response to this statement, we suggest a new organisation of the necessary models, stating clearly the links and separation between models and planning algorithms. This has lead to the development of a C++ library that structures the available models and defines the requests required by the planning process. We then exploit this library through a set of algorithms tackling area patrolling and target tracking. These algorithms are supported by a sound formalism and we study the impact of the models on the observed performances, with an emphasis on the complexity and the quality of the resultingsolutions. As a more general consideration, models are an essential link between Artificial Intelligence and applied Robotics : improving their expressiveness and studying them rigorously are the keys leading toward better robot behaviours and successful robotic missions. This thesis help to show how important the models are for planning and other decision processes formulti-robot missions.

Due to recent improvements, robots are more applicable in factories and various production lines where smoke, fog, dust, and steam are inevitable. Despite their advantages, robots introduce new safety requirements when combined with humans. Radars can play a crucial role in this context by providing safe zones where robots are operating in the absence of humans. The goal of this Master’s thesis is to investigate different clutter suppression methods for single radar sensor reflection data via digital signal processing. This was done in collaboration with ABB Jokab AB, Sweden. The calculations and implementation of the digital signal processing algorithms are made with Octave. A critical problem is false detection that could possibly cause irreparable damage. Therefore, a safety system with an extremely low false alarm rate is desired to reduce costs and damages. In this project, we have studied four different digital low pass filters: moving average, multiple-pass moving average, Butterworth, and window-based filters. The results are compared, and it is ascertained that all the results are logically compatible, broadly comparable, and usable in this context.

No description available.

<p>Deep learning has been increasingly adopted in a multitude of settings. Yet, its strong performance relies on processing data during inference that is in-distribution with its training data. Deep learning input data during deployment, however, is not guaranteed to be in-distribution with the model's training data and can often times be distorted, either intentionally (e.g., by an adversary) or unintentionally (e.g., by a sensor defect), leading to significant performance degradations. In this dissertation, we develop algorithms for a variety of applications to improve the performance of deep learning models in the presence of distorted data. We begin by first designing feature engineering methodologies to increase classification performance in noisy environments. Here, we demonstrate the efficacy of our proposed algorithms on two target detection tasks and show that our framework outperforms a variety of state-of-the-art baselines. Next, we develop mitigation algorithms to improve the performance of deep learning in the presence of adversarial attacks and nonlinear signal distortions. In this context, we demonstrate the effectiveness of our methods on a variety of wireless communications tasks including automatic modulation classification, power allocation in massive MIMO networks, and signal detection. Finally, we develop an uncertainty quantification framework, which produces distributive estimates, as opposed to point predictions, from deep learning models in order to characterize samples with uncertain predictions as well as samples that are out-of-distribution from the model's training data. Our uncertainty quantification framework is carried out on a hyperspectral image target detection task as well as on counter unmanned aircraft systems (cUAS) model. Ultimately, our proposed algorithms improve the performance of deep learning in several environments in which the data during inference has been distorted to be out-of-distribution from the training data. </p>

No description available.

The main aim of using radars in industrial safety system is to detect the presence of target accurately. The conventional methods of radar target detection algorithm such as the Cell averaging constant false alarm rate method (CA-CFAR), Greatest of constant false alarm method (GO-CFAR) and the Smallest of constant false alarm rate method (SO-CFAR) has their own disadvantage when it comes to precise target detection which is a key factor for a safety system. This thesis investigates the above mentioned conventional CFAR algorithms for its pros and cons in target detection and proposes a new and improved method called Cell Evaluation target detection method. The proposed method is shown to mitigate the limitations present and the assumptions made in the conventional target detection method. Further more angular estimation is performed to determine the precise location of the target and the artifacts due to the angular estimation is eliminated by aggregating the detected points from multiple radar modules by linear translation. This gives a better visualization of the target. / Radarteknik kan användas inom maskinsäkerhet (MS) för att detektera skyddsvärda objekt, typiskt människor i arbete nära maskiner. Konventionella metoder för detektering med given frekvens falsk alarm (eng. Constant False Alarm Rate (CFAR)) som baseras på medelvärden, har dock betydande brister. Främst beträffande precision och tillförlitlighet, vilket är centralt för MS. Exempel som studerats i detta examensarbete är “Cell-averaging CFAR” (CA-CFAR), “Greatest of CFAR” (GO-CFAR) samt “Smallest of CFAR” (SO-CFAR). Med målet att förbättra detektionen föreslås även en ny CFAR-metod, vilken benämns ”Cell Evaluation target detection”. I detta arbete visas denna metod undertrycka begränsningar med konventionella tekniker. Den undviker även en del antaganden som inte alltid stämmer i praktiken. Studien inkluderar även skattning av riktning. Det visas hur visualisering av skyddsobjekt kan förbättras, genom att felaktigheter elimineras efter sammanläggning av detektioner från flera radarmoduler efter koordinattransformation.

<p>Small, low cost Autonomous underwater vehicles (AUVs) provide ideal platforms for shallow water survey, as they are capable of unmanned navigation and can be programmed to acquire data at constant depth, or constant altitude above the seabed. AUVs can be deployed under most sea states and are unaffected by vessel motions that often degrade sonar and magnetometer data quality. The integration of sonar and magnetometer sensors on AUV’s is challenging, however, due to limited payload and strong magnetic fields produced by the vehicle motor.</p> <p>In this study, a Marine Magnetics Explorer Overhauser magnetometer was mated to a portable AUV (OceanServer Iver2) creating the first practical AUV- deployed magnetic survey system. To eliminate magnetic interference from the AUV, the magnetometer was tethered to the AUV with a 5 m tow cable, as determined by static and dynamic instrument testing. The results of the magnetic tests are presented, along with field data from a shallow water test area in Lake Ontario near Toronto, Canada. AUV-acquired magnetic survey data were compared directly with a conventional boat-towed magnetic survey of the same area. The AUV magnetic data were of superior quality despite being collected in rough weather conditions that would have made conventional survey impossible. The resulting high-resolution total magnetic intensity and analytic signal maps clearly identify several buried and surface ferrometallic targets that were verified in 500-kHz side- scan sonar imaging and visual inspection by divers.</p> / Master of Science (MSc)

L’imagerie hyperspectrale, qui consiste à acquérir l'image d'une scène dans un grand nombre de bandes spectrales, permet de détecter des cibles là où l'imagerie couleur classique ne permettrait pas de conclure. Les imageurs hyperspectraux à acquisition séquentielle sont inadaptés aux applications de détection en temps réel. Dans cette thèse, nous proposons d’utiliser un imageur multispectral snapshot, capable d’acquérir simultanément un nombre réduit de bandes spectrales sur un unique détecteur matriciel. Le capteur offrant un nombre de pixels limité, il est nécessaire de réaliser un compromis en choisissant soigneusement le nombre et les profils spectraux des filtres de l'imageur afin d’optimiser la performance de détection. Dans cet objectif, nous avons développé une méthode de sélection de bandes qui peut être utilisée dans la conception d’imageurs multispectraux basés sur une matrice de filtres fixes ou accordables. Nous montrons, à partir d'images hyperspectrales issues de différentes campagnes de mesure, que la sélection des bandes spectrales à acquérir peut conduire à des imageurs multispectraux capables de détecter des cibles ou des anomalies avec une efficacité de détection proche de celle obtenue avec une résolution hyperspectrale. Nous développons conjointement un démonstrateur constitué d'une matrice de 4 filtres de Fabry-Perot accordables électroniquement en vue de son implantation sur un imageur multispectral snapshot agile. Ces filtres sont développés en technologie MOEMS (microsystèmes opto-électro-mécaniques) en partenariat avec l'Institut d'Electronique Fondamentale. Nous présentons le dimensionnement optique du dispositif ainsi qu'une étude de tolérancement qui a permis de valider sa faisabilité. / Hyperspectral imaging, which consists in acquiring the image of a scene in a large number of spectral bands, can be used to detect targets that are not visible using conventional color imaging. Hyperspectral imagers based on sequential acquisition are unsuitable for real-time detection applications. In this thesis, we propose to use a snapshot multispectral imager able to acquire simultaneously a small number of spectral bands on a single image sensor. As the sensor offers a limited number of pixels, it is necessary to achieve a trade-off by carefully choosing the number and the spectral profiles of the imager’s filters in order to optimize the detection performance. For this purpose, we developed a band selection method that can be used to design multispectral imagers based on arrays of fixed or tunable filters. We use real hyperspectral images to show that the selection of spectral bands can lead to multispectral imagers able to compete against hyperspectral imagers for target detection and anomaly detection applications while allowing snapshot acquisition and real-time detection. We jointly develop an adaptive snapshot multispectral imager based on an array of 4 electronically tunable Fabry-Perot filters. The filters are developed in MOEMS technology (Micro-Opto-Electro-Mechanical Systems) in partnership with the Institut d'Electronique Fondamentale. We present the optical design of the device and a study of tolerancing which has validated its feasibility.