Pedestrian movement is unconstrained. For this reason it is not amenable to mathematical modelling in the same way as road trac. Individual pedestrians are notoriously difficult to monitor at a microscopic level. This has led to a lack of primary data that can be used to develop reliable models. Although video surveillance is cheap to install and operate, video data is extremely expensive to process for this purpose. An alternative approach is to use passive infrared detectors that are able to track individuals unobtrusively. This thesis describesthe use of a low cost infrared sensor for use in tracking pedestrians. The sensor itself, manufactured by a British company, is designed to count people crossing an arbitrary datum line. However, with the development of additional software, the functionality of these sensors can be extended beyond their original design specication. This allows the trajectories of individual pedestrians to be tracked. Although the field of view of each sensor is relatively small (44 m), five were deployed in a busy indoor corridor, covering most of its length. In this research, the technical challenges involved in using the sensors in this way are addressed. Statistics derived from the data collected are then compared to other studies at this scale.

Cette thèse présente l'utilisation conjointe de la théorie de l'évidente et du suivi multi-hypothèses pour la modélisation et la gestion d'un système de suivi multi-caméras dans un environnement autoroutier. Le suivi est basé sur la ré-identification des objets (véhicules) sur la base d'information visio-temporelles. Une concrétisation de ces concepts se traduit par la conception et la mise en oeuvre d'une architecture logicielle multi-agents de gestion du suivi multi-caméras. Après une présentation de l'état de l'art sur les cadres de gestion de l'incertain et celui relatif à fusion de l'information pour la mise en correspondance, et sur les systèmes multi-agents, notre apport dans ce travail se situe à trois niveaux. Le premier a été une adaptation de la phase de décision du modèle de croyance transférable pour y intégrer l'utilisation du suivi multi-hypothèses comme outil de levée d'ambigüité rn cas d'indécision face à une situation de mise en correspondance. Le second apport a été celui de proposer une architecture logicielle à base d'agents pour la gestion du système du suivi multi-caméras. Nous en avons proposé la modélisation globale ainsi que celle des agents et de leurs interactions en utilisant une démarche personnelle d'analyse mais toutefois inspirée de langages et outils de modélisation tels que Agent UML et MaSE pour ne citer que ceux-là, du fait qu'il n'existe pas réellement un standard normalisé à ce jour dans ce domaine. Notre troisième apport a été de faire un début d'implémentation de notre architecture logicielle à base d'agent en nous basant sur la plateforme JADE (Java Agent DEvelopment Framework). Quelques expérimentations et discussions des résultats sont présentées à la fin pour déboucher sur nos conclusions et perspectives. / This thesis presents the joint use of the theory of evidence and multiple hypothesis tracking for modeling and managing a system for monitoring multiple cameras in a motorway. The tracking is based on the re-identification of objects (vehicles) on the basis of visuals and times informations. A realization of these concepts results in the design and implementation of a software architecture for multiple agents management of multiple camera tracking system. After presenting the state of the art on the frameworks of uncertainty management and that on information fusion for the matching, and the multi-agent systems, our contribution in this work is on two or three levels. The first was an adaptation of the decision phase of the transferable belief model to incorporate the use of multi-hypotheses tracking as a tool of ambiguity survey in case of indecision in matching situation. The second contribution was a proposition of agent-based software architecture for management of a multiple cameras tracking system. We have proposed the global system modeling as well as agents and their interactions modeling using a personal analysis method but nevertheless inspired by modelisation languages and tolls such as Agent UML, MaSE and others, because there is not yet a standard and normalized tool on the subject. Our third contribution was to begin an implementation of our agent-based software architecture using JADE (Java Agent Development Framework). Some experiment and discussions are presented at the end to lead to our conclusions and perspectives.

La vision artificielle est un domaine de recherche en pleine évolution. Les nouvelles stratégies permettent d'avoir des réseaux de caméras intelligentes. Cela induit le développement de beaucoup d'applications de surveillance automatique via les caméras. Les travaux développés dans cette thèse concernent la mise en place d'un système de vidéosurveillance intelligente pour la détection de chutes en temps réel. La première partie de nos travaux consiste à pouvoir estimer de façon robuste la surface d'une personne à partir de deux (02) caméras ayant des vues complémentaires. Cette estimation est issue de la détection de chaque caméra. Dans l'optique d'avoir une détection robuste, nous avons fait recours à deux approches. La première approche consiste à combiner un algorithme de détection de mouvements basé sur la modélisation de l'arrière plan avec un algorithme de détection de contours. Une approche de fusion a été proposée pour rendre beaucoup plus efficiente le résultat de la détection. La seconde approche est basée sur les régions homogènes de l'image. Une première ségmentation est effectuée dans le but de déterminer les régions homogènes de l'image. Et pour finir, nous faisons la modélisation de l'arrière plan en se basant sur les régions. Une fois les pixels de premier plan obtenu, nous faisons une approximation par un polygone dans le but de réduire le nombre d'informations à manipuler. Pour l'estimation de cette surface nous avons proposé une stratégie de fusion dans le but d'agréger les détections des caméras. Cette stratégie conduit à déterminer l'intersection de la projection des divers polygones dans le plan de masse. La projection est basée sur les principes de l'homographie planaire. Une fois l'estimation obtenue, nous avons proposé une stratégie pour détecter les chutes de personnes. Notre approche permet aussi d'avoir une information précise sur les différentes postures de l'individu. Les divers algorithmes proposés ont été implémentés et testés sur des banques de données publiques dans le but de juger l'efficacité des approches proposées par rapport aux approches existantes dans l'état de l'art. Les résultats obtenus et qui ont été détaillés dans le présent manuscrit montrent l'apport de nos algorithmes. / Artificial vision is an involving field of research. The new strategies make it possible to have some autonomous networks of cameras. This leads to the development of many automatic surveillance applications using the cameras. The work developed in this thesis concerns the setting up of an intelligent video surveillance system for real-time people fall detection. The first part of our work consists of a robust estimation of the surface area of a person from two (02) cameras with complementary views. This estimation is based on the detection of each camera. In order to have a robust detection, we propose two approaches. The first approach consists in combining a motion detection algorithm based on the background modeling with an edge detection algorithm. A fusion approach has been proposed to make much more efficient the results of the detection. The second approach is based on the homogeneous regions of the image. A first segmentation is performed to find homogeneous regions of the image. And finally we model the background using obtained regions.

El objetivo de la presente tesina es el estudio técnico para la Autoridad del Servicio Civil - Lima, orientado a diseñar un sistema de Videovigilancia IP, lo cual es una tecnología de vigilancia visual con las ventajas digitales de las redes de comunicación IP (Internet Protocol), lo cual permite la supervisión local y/o remota de imágenes y audio así como el tratamiento digital de las imágenes. Existe una gama muy amplia de cámaras con diferentes funcionalidades para satisfacer todas las necesidades y presupuestos. El objetivo es diseñar un sistema de Videovigilancia IP, el cual permita vigilar y controlar a todo el personal para así poder disminuir las pérdidas de los activos. 1. Diseñar un sistema de Videovigilancia IP para una supervisión continua de los datos enviados por la cámara IP. 2. Diseñar la red lógica y física de la Videovigilancia IP. 3. Propuesta de un plan de capacitación de personal involucrado en el manejo de los equipos de Videovigilancia IP, alcanzados en este trabajo. The aim of this thesis is the technical study for the Civil Service Authority - Lima, guied at designing an IP video-surveillance system, which is a visual surveillance technology with digital advantages of IP communication networks (Internet Protocol), allowing the local supervision and/or remote images and audio; and the digital image processing. There is a wide range of cameras with different features to satisfy all needs and budgets. The goal is to design an IP video-surveillance system, which allows monitoring and controlling all personnel in order to reduce losses of assets. 1. Design an IP video-surveillance system for continuous monitoring of the data sent by the IP camera. 2. Design the logical and physical network of video-surveillance IP. 3. Proposal of a training plan for staff involved in the management of IP video-surveillance equipment, made in this work.

A violência urbana e o medo globalizado, marcas do atual período técnico-científico e informacional, têm alterado as paisagens de diversas partes do mundo através do processo denominado neste trabalho pelo termo securização urbana. Como resposta à sensação de insegurança e de imprevisibilidade, busca-se uma racionalização do território a partir da informatização do cotidiano e da criação de espaços exclusivos. Empiricamente, a securização se traduz em formas arquitetônicas variadas, tendo sido destacadas nesta tese as câmeras de vigilância, os condomínios fechados e as arquiteturas anti-indesejáveis. Transformações espaciais como essas são particularmente intensas em Campinas-SP, município brasileiro ao mesmo tempo muito rico, com importantes empresas e universidades, e muito pobre e violento, portando índices de criminalidade acima da média nacional. Exemplos desse e também de outros lugares do Brasil e da Europa foram analisados em trabalhos de campo que contaram com entrevistas colhidas de agentes locais, fotografias e cartografias, com o intuito de caracterizar o processo de securização e, especialmente, conduzir a uma reflexão sobre suas conseqüências. Conclui-se que a maneira pela qual a segurança tem sido buscada aumenta as desigualdades espaciais e promove uma privatização dos espaços públicos. Além disso, o excesso de vigilância tem reforçado a psicoesfera do medo, tolhido muitas das liberdades individuais e criado novas neuroses e violências. A racionalização do espaço para fins de segurança cria, contudo, as condições para o surgimento de contra-racionalidades, o que reafirma o caráter complexo e dialético do espaço geográfico e aponta para a possibilidade de um futuro marcado pelas solidariedades geográficas e pelo poder revolucionário dos agentes não-hegemônicos. / Urban violence and globalized fear, hallmarks of the current technical-scientific and informational period, have transformed the landscape of different cities of the world through a process called urban securitization\". In response to the feeling of insecurity and unpredictability, territories are being rationalised through the digitalization of everyday life and creation of exclusive areas. Empirically, securitization can be materialized through different architectural forms. This thesis highlights surveillance cameras, gated communities and anti-beggars architectures. These transformations are particularly intense in Campinas, a very wealthy Brazilian city, with important companies and universities, and at the same time very poor and violent, with crime rates above the national average. Examples from this city, but also from other parts of Brazil and Europe, were analyzed in field works which involved interviews with local agents, photos and maps, in order to describe the process of securitization and, especially, to lead to a reflection of their consequences. In conclusion, it can be stated that the way security is being searched increases spatial inequalities and promotes a privatization of public spaces. Furthermore, an excessive surveillance has enhanced the psycho-sphere of fear, has restraint individual liberties and has produced new neuroses and violence. However, rationalisation of space for security purposes facilitates the emergence of counter-rationalities, emphasizing the complex and dialectic qualities of geographic space and indicating the possibility of a future characterized by geographic solidarities and by revolutionary power of non-hegemonic agents.

The problem of face detection and recognition within a given database has become one of the important problems in computer vision. A simple approach for Face Detection in video is to run a learning based face detector every frame. But such an approach is computationally expensive and completely ignores the temporal continuity present in videos. Moreover the search space can be reduced by utilizing visual cues extracted based on the relevant task at hand(top down approach). Once detection is done next step is to perform a face recognition based on the available database. But the faces detected from face detect or output is neither aligned nor well cropped and is prone to scale change. We call such faces as free form faces. But the current existing algorithms on face recognition assume faces to be properly aligned and cropped, and having the same scale as the faces in the database, which is highly constrained. In this thesis, we propose an integrated detect-track framework for Multiview face detection in videos. We overcome the limitations of the frame based approaches, by utilizing the temporal continuity present in videos and also incorporating the top down information of the task. We model the problem based on the concept from Experiential sampling [2]. This consists of determining certain key positions which are relevant to the task(face detection). These key positions are referred to as attention samples and Multiview face detection is performed only at these locations. These statistical samples are estimated based on the visual cues, past experience and the temporal continuity and is modeled as a Bayesian filtering problem, which is solved using Particle Filters. In order to detect all views we use a tracker integrated with the detector and come out with a novel track termination algorithm using the concepts from Track Before Detect(TBD)[26]. Such an approach is computationally efficient and also results in lower false positive rate. We provide experiments showing the efficiency of the integrated detect-track approach over the multiview face detector approach without a tracker. For free form face recognition we propose to use the concept of Principal Geodesic Analysis(PGA) of the Covariance descriptors obtained from Gabor filters. This is similar to Principal Component Analysis in Euclidean spaces (Covariance descriptors lie on a Riemannian manifold). Such a descriptor is robust to alignment and scaling problems and also are of lower dimensions. We also employ sparse modeling technique for Face recognition task using these Covariance descriptor which are dimensionally reduced by transforming them on to a tangent space, which we call PGA feature. Further, we improve upon the recognition results of linear sparse modeling, by non-linear mapping of the PGA features by employing “Kernel Trick” for these sparse models. We show that the Kernelized sparse models using the PGA features are indeed very efficient for free form face recognition by testing on two standard databases namely AR and YaleB database.

Práca rieši možnosť zaznamenávať pozíciu osôb v zázname z kamery drona a určovať ich polohu. Absolútna pozícia sledovanej osoby je odvodená vzhľadom k pozícii kamery, teda vzhľadom k umiestneniu drona vybaveného príslušnými senzormi. Zistené dáta sú po ich spracovaní vykreslené ako príslušné cesty v grafe. Práca si ďalej dáva za cieľ využiť dostupné riešenia čiastkových problémov: detekcia osôb v obraze, identifikácia jednotlivých osôb v čase, určenie vzdialenosti objektu od kamery, spracovanie potrebných senzorových dát. Následne využiť preskúmané metódy a navrhnúť riešenie, ktoré bude v reálnom čase pracovať na uvedenom probléme. Implementačná časť spočíva vo využití akcelerátoru Intel NCS v spojení s Raspberry Pi priamo ako súčasť drona. Výsledný systém je schopný generovať výstup o polohe detekovaných osôb v zábere kamery a príslušne ho prezentovať.

Die vorliegende Arbeit basiert auf der Sekundäranalyse eines Datensatzes, welcher im Stadtgebiet Leipzig im Rahmen eines Forschungspraktikums erhoben wurde. Die Autoren folgen modernen sozialwissenschaftlichen Standards, indem sie theoretisch abgeleitete Hypothesen empirisch prüfen. Besonders hervorhebenswert ist dabei, dass sie ein komplexes Mehrebenenmodell zur Erklärung von Einstellungen zum Überwachen und Strafen theoretisch versiert entwickeln. Beides, Überwachen und Strafen, sind hochbrisante aktuelle Themen in der öffentlichen Diskussion. Gerade in diesen sensiblen Sicherheitsbereichen erweist sich die Bevölkerungsmeinung als ausgesprochen instabil. Immer wieder gelingt es über Mediendramaturgien beachtliche Teile der öffentlichen Meinung für die populistische Konstruktion der Erhöhung der Strafhärte zu mobilisieren. Beeinflussungen der Kriminalitätsfurcht erweisen sich dabei als besonders effektiv. Über die affektive Komponente gelingt es nicht selten auch die kognitive, die Urteilsbildung zu beeinflussen. Hinsichtlich der Einstellungen zum Überwachen ist die Situation ebenfalls nicht besser. Zumindest kurzfristig haben demoskopische Untersuchungen gezeigt, dass ein gewisser Teil der Bevölkerung bereit ist, demokratische Rechte zur Erhöhung der öffentlichen Sicherheit wenn nicht aufzugeben so doch zumindest merklich einschränken zu lassen.:Einleitung/Fragestellung der Untersuchung; Theorien zum Einstellungskonzept, Theoretischer und empirischer Forschungsstand, Auswertung der Leipziger Kriminalitätsstudie, Zusammenfassung und Schlußbetrachtung

Multiple target tracking (MTT) is the process of identifying the number of targets present in a surveillance region and the state estimates, or track, of each target. MTT remains a challenging problem due to the NP-hard data association step, where unlabeled measurements are identified as either a measurement of an existing target, a new target, or a spurious measurement called clutter. Existing techniques suffer from at least one of the following drawbacks: divergence in clutter, underlying assumptions on the number of targets, high computational complexity, time-consuming implementation, poor performance at low detection rates, and/or poor track continuity. Our goal is to develop an efficient MTT algorithm that is simple yet effective and that maintains track continuity enabling persistent tracking of an unknown number of targets. A related field to tracking is regression analysis, where the parameters of static signals are estimated from a batch or a sequence of data. The random sample consensus (RANSAC) algorithm was developed to mitigate the effects of spurious measurements, and has since found wide application within the computer vision community due to its robustness and efficiency. The main concept of RANSAC is to form numerous simple hypotheses from a batch of data and identify the hypothesis with the most supporting measurements. Unfortunately, RANSAC is not designed to track multiple targets using sequential measurements.To this end, we have developed the recursive-RANSAC (R-RANSAC) algorithm, which tracks multiple signals in clutter without requiring prior knowledge of the number of existing signals. The basic premise of the R-RANSAC algorithm is to store a set of RANSAC hypotheses between time steps. New measurements are used to either update existing hypotheses or generate new hypotheses using RANSAC. Storing multiple hypotheses enables R-RANSAC to track multiple targets. Good tracks are identified when a sufficient number of measurements support a hypothesis track. The complexity of R-RANSAC is shown to be squared in the number of measurements and stored tracks, and under moderate assumptions R-RANSAC converges in mean to the true states. We apply R-RANSAC to a variety of simulation, camera, and radar tracking examples.

The purpose of this review of literature is to investigate the various roles of video surveillance (VS) in the hospital's intensive care unit (ICU) as well as its legal and ethical implications. Today, hospitals spend more money on the ICU than on any other unit. By 2030, the population of those 65 and over is expected to double. 80% of older adults have at least one chronic diseases (Centers for Disease Control and Prevention, 2013). As a consequence, the demand for ICU services will likely increase, which may burden hospital with additional costs. Because of increasing economic pressures, more hospitals are using video surveillance to enhance quality care and reduce ICU costs (Goran, 2012). Research shows that VS enhances positive outcomes among patients and best practice compliance among hospital staff. The results are fewer reports of patient complications and days spent in the ICU, and an increase in reported hospital savings. In addition, VS is becoming an important tool for the families of newborns in the neonatal ICU (NICU). The belief is that the VS can facilitate parent-baby bonding. In the United States of America, privacy rights impose legal restrictions on VS. These rights come from the U.S. Constitution, Statutory law, Regulatory law, and State law. HIPPA authorizes the patient to control the use and disclosure of his or her health information. Accordingly, hospitals are under obligation to inform patients on their right to protected health information. It is appropriate that hospitals use VS for diagnostic purposes as long as they have obtained patient consent. According to modern day privacy experts Charles Fried and Alan Westin, a violation of a person's privacy equates a violation on their liberty and morality. However, if a physician suspects that a third party person is causing harm to the patient, than the use of covert VS is justifiable.