In multi-camera surveillance, association of people detected in different camera views over time, known as person re-identification, is a fundamental task. Re-identification is a challenging problem because of changes in the appearance of people under varying camera conditions. Existing approaches focus on improving the re-identification accuracy, while no specific effort has yet been put into efficiently utilising the available resources that are normally limited in a camera network, such as storage, computation and communication capabilities. In this thesis, we aim to perform and improve the task of re-identification under constrained resources. More specifically, we reduce the data needed to represent the appearance of an object through a proposed feature selection method and a difference-vector representation method. The proposed feature-selection method considers the computational cost of feature extraction and the cost of storing the feature descriptor jointly with the feature's re-identification performance to select the most cost-effective and well-performing features. This selection allows us to improve inter-camera re-identification while reducing storage and computation requirements within each camera. The selected features are ranked in the order of effectiveness, which enable a further reduction by dropping the least effective features when application constraints require this conformity. We also reduce the communication overhead in the camera network by transferring only a difference vector, obtained from the extracted features of an object and the reference features within a camera, as an object representation for the association. In order to reduce the number of possible matches per association, we group the objects appearing within a defined time-interval in un-calibrated camera pairs. Such a grouping improves the re-identification, since only those objects that appear within the same time-interval in a camera pair are needed to be associated. For temporal alignment of cameras, we exploit differences between the frame numbers of the detected objects in a camera pair. Finally, in contrast to pairwise camera associations used in literature, we propose a many-to-one camera association method for re-identification, where multiple cameras can be candidates for having generated the previous detections of an object. We obtain camera-invariant matching scores from the scores obtained using the pairwise re-identification approaches. These scores measure the chances of a correct match between the objects detected in a group of cameras. Experimental results on publicly available and in-lab multi-camera image and video datasets show that the proposed methods successfully reduce storage, computation and communication requirements while improving the re-identification rate compared to existing re-identification approaches.

Object appearance models are a consequence of illumination, viewing direction, camera intrinsics, and other conditions that are specific to a particular camera. As a result, a model acquired in one view is often inappropriate for use in other viewpoints. In this work we treat this appearance model distortion between two non-overlapping cameras as one in which some unknown color transfer function warps a known appearance model from one view to another. We demonstrate how to recover this function in the case where the distortion function is approximated as general affine and object appearance is represented as a mixture of Gaussians. Appearance models are brought into correspondence by searching for a bijection function that best minimizes an entropic metric for model dissimilarity. These correspondences lead to a solution for the transfer function that brings the parameters of the models into alignment in the UV chromaticity plane. Finally, a set of these transfer functions acquired from a collection of object pairs are generalized to a single camera-pair-specific transfer function via robust fitting. We demonstrate the method in the context of a video surveillance network and show that recognition of subjects in disjoint views can be significantly improved using the new color transfer approach.

THE MONITORING EMPLOYEES AT WORK The thesis deals with the legal protection of privacy and property in connection with the employees monitoring. The legal practice solves this problematic area in a different ways, which makes this topic highly actual. The thesis is composed of six chapters, each of them describes important aspects of the employees monitoring. First two chapters are introductory and define basic terminology and legal principles related to the personal data protection. Chapter Three focuses on the privacy as a fundamental human right and points out the issue of the confrontation with other values. Possibilities to enforce the right of privacy are outlined in the last part of this chapter. Chapter Four examines relevant Czech legislation concerning employees monitoring and personal data protection in labor law relations. The means of employees monitoring are analyzed in Chapter Five, which is subdivided into five parts - camera surveillance systems, correspondence confidentiality, internet use, GPS localization and biometric identification and authentication. Chapter Six seeks to describe the relevant Czech case law and illustrates the approach to decision-making by the administrative authority. The thesis strives for the brief analysis of employees monitoring in terms of generally binding...

The integration of image sensors and high performance processors into embedded systems enabled the development of intelligent vision systems. In this thesis, we developed an active autonomous system to be used for surveillance applications. The proposed system detects a single moving object in the field of view automatically and tracks it in a wide area by controlling the pan-tilt-zoom features of the camera. The system can also go to an alarm state to warn the user. The processing unit of the system is a Texas Instruments DM642 Evaluation Module which is a low-cost high performance video &amp / imaging development platform designed to develop and evaluate video based applications.

Looking at opportunities and difficulties with the use of camera surveillance is important for the futureof this technology and its use by the Swedish police. The purpose of this assignment is therefore toprovide an overview of the opportunities and difficulties encountered by the Swedish police in the useof camera surveillance in their daily work. The society is affected by these cameras, but what does thesociety think about being monitored in public places? Through interviews and a backgroundpresentation of camera surveillance as well as a theoretical approach with a focus on moralizingtechnology, we have analyzed and discussed various opportunities and problems that the policeencounter. According to studies, a large majority is in favor of camera surveillance, yet there are strictlaws and regulations governing their use. Replacing and simplifying laws will bring responsibility anddecisions closer to the police, opening up possibilities such as the strengthening of witness statementsand a better overview of crime sites, while fear that the technology does not reflect the reality of thesituation. The material provided by the camera surveillance also provides an objective overview whichwould not otherwise have been possible, while police surveillance has limitations. / Att titta på möjligheter och svårigheter med användandet av kamerabevakning är viktigt för framtideninom denna teknologi och dess användning av polisen. Syftet med denna uppsats är att ge en överblickav de möjligheter och svårigheter som svenska polisen stöter på gällande användandet avkamerabevakning i sitt arbete. Allmänheten påverkas av dessa kameror, men vad tycker egentligenallmänheten om att bli övervakad på allmänna platser? Genom en presentation av förutsättningar förkamerabevakning och intervjuer samt ett teoretiskt synsätt med fokus på moralisering av teknologi, harvi analyserat och diskuterat olika möjligheter och problem som polisen stöter på. Enligt andra studierär en stor majoritet positivt inställda till kamerabevakning, trots detta finns det stränga lagar och reglergällande användning av dessa. I samband med att lagar ersätts och förenklas flyttas ansvaret ochbesluten närmare polisens verksamhet vilket öppnat upp för möjligheter som bland annat stärkandet avvittnesutsagor och en bättre överblick av brottsplatser samtidigt som en rädsla finns för att teknikeninte speglar den verklighetsbild som är korrekt. Materialet som kamerabevakningen tillför ger även enobjektiv överblick som annars inte hade varit möjlig, samtidigt som polisens kamerabevakning harbegränsningar.

With increased capabilities in technology we’ve seen a huge societal impact in how we live our lives and what can be done with our personal information. Even if the purpose of camera surveillance might be done with good intent, by surveilling a population the question of privacy comes to light. The study aims to understand how camera surveillance affects privacy. The study also looks forward into the future with the possibilities of facial recognition and how this technology would change camera surveillance and in turn privacy. Important to note is that this study does not aim to examine whether camera surveillance is an effective means of crime prevention or crime investigation, but rather the consequences it entails for people’s privacy and how the infringement on privacy is seen from an ethical and a judicial standpoint. This study’s focus is examining camera surveillance and privacy in Sweden. The results and conclusion of the study derives from 4 different themes that were identified during the method segment, these are, Ethics, Law, Camera Surveillance, and Facial Recognition. Firstly, the study argues that trust and transparency needed for camera surveillance to be accepted by the public and to make sure that the camera surveillance is ethical. Secondly, we argue there is an ongoing infringement on privacy and additional risks if the storing of personal information gathered by surveillance cameras exists. Thirdly, an imbalance between technological possibilities and legal possibilities can lead to either a threat- or protection for privacy. Furthermore, the study acknowledges the importance of government agencies to be able to uphold laws and regulations regarding camera surveillance and privacy. Finally, the study finds mixed views regarding camera surveillance mixed with additional technologies, such as facial recognition. To increase trust and to make sure the technology serves the people we argue that some sort of ethical governance is needed.

In this thesis, novel methods for background modeling, tracking, occlusion handling and event recognition via multi-camera configurations are presented. As the initial step, building blocks of typical single camera surveillance systems that are moving object detection, tracking and event recognition, are discussed and various widely accepted methods for these building blocks are tested to asses on their performance. Next, for the multi-camera surveillance systems, background modeling, occlusion handling, tracking and event recognition for two-camera configurations are examined. Various foreground detection methods are discussed and a background modeling algorithm, which is based on multi-variate mixture of Gaussians, is proposed. During occlusion handling studies, a novel method for segmenting the occluded objects is proposed, in which a top-view of the scene, free of occlusions, is generated from multi-view data. The experiments indicate that the occlusion handling algorithm operates successfully on various test data. A novel tracking method by using multi-camera configurations is also proposed. The main idea of multi-camera employment is fusing the 2D information coming from the cameras to obtain a 3D information for better occlusion handling and seamless tracking. The proposed algorithm is tested on different data sets and it shows clear improvement over single camera tracker. Finally, multi-camera trajectories of objects are classified by proposed multi-camera event recognition method. In this method, concatenated different view trajectories are used to train Gaussian Mixture Hidden Markov Models. The experimental results indicate an improvement for the multi-camera event recognition performance over the event recognition by using single camera.

For many years, camera surveillance has been used for crime prevention purposes in public places, but also in vulnerable areas. However, the use of camera surveillance has in many cases been complicated in connection with, among other things; privacy, exercise of power and segregation. This is a study on camera surveillance in vulnerable areas with a demarcation point to the city of Uppsala and Västerås. The purpose of this study is to see how the inhabitants of the vulnerable areas, actually experience camera surveillance in vulnerable areas. The study also aims to see what intentions are behind camera surveillance in the vulnerable areas of Uppsala and Västerås. The choice of method is based on a qualitative study in the form of semi-structured interviews with a total of ten respondents, eight of whom are residents from vulnerable areas in Uppsala and Västerås and two of the respondents are municipal employees from Uppsala and Västerås municipality. The two respondents with municipal employment have different years of experience in surveillance, security and crime prevention. The data analysis has been in the form of both an inductive and deductive analysis where different themes have been used to clarify the results. The selection of the different themes has been based on both the theoretical framework and the collected data. It turned out that for the research question; “How is camera surveillance in vulnerable areas perceived by residents from vulnerable areas in Uppsala and Västerås?” there were varied responses in certain occasions, for different categories. Some of the respondents saw the good of camera surveillance, while others saw it as an offensive and sometimes unnecessary measure. For the supplementary side-to-side question; “What are the intentions behind camera surveillance in vulnerable areas in Uppsala and Västerås?" the two municipal respondents had similar answers; camera surveillance aimed to reduce crime, especially drug sales, which both respondents mentioned. / Kameraövervakning har i många år använts i brottsförebyggande syfte på allmänna platser, men även i utsatta områden. Dock har användningen av kameraövervakningen i många fall varit komplicerad i samband med bland annat den personliga integriteten, maktutövning och segregation. Därmed handlar föreliggande studie om kameraövervakningen i utsatta områden med avgränsning till Uppsala och Västerås stad. Syftet med föreliggande studie är att se hur invånarna i de utsatta områdena, faktiskt upplever kameraövervakningen i utsatta områden. Studien syftar även till att se vilka intentioner som ligger bakom kameraövervakningen i de utsatta områdena i Uppsala och Västerås. Valet av metod grundar sig på en kvalitativ studie i form av semistrukturerade intervjuer med totalt tio respondenter, varav åtta är invånare från utsatta områden i Uppsala och Västerås och två av respondenterna är kommunanställda från Uppsala- och Västerås kommun. De två respondenterna med kommunanställning har olika års erfarenhet kring just övervakning, säkerhet och det brottsförebyggande arbetet. Dataanalysen har varit i form av både en induktiv samt deduktiv analys där olika teman har använts för att tydliggöra resultatet. Urvalet för de olika teman har grundat sig på dels det teoretiska ramverket och dels det insamlade datamaterialet. Det visade sig att för forskningsfrågan; ”Hur upplevs kameraövervakningen i utsatta områden av invånare från Uppsala- och Västerås utsatta områden?”  varierade svaren i vissa tillfällen, för olika kategorier. Vissa av respondenterna såg det goda med kameraövervakningen, medan andra såg det som en kränkande och ibland onödig åtgärd. För den kompletterande bifrågan; ”Vad är intentionerna bakom kameraövervakningen i utsatta områden i Uppsala och Västerås?” gav de två kommunanställda respondenterna liknande svar; kameraövervakningen syftade till att minska på brottsligheten, framförallt narkotikaförsäljning, som båda respondenterna tog upp.

The aim of this study is to examine how students and personnel at universities perceive camera surveillance and the handling of the data gathered by it. This is done through a qualitative research approach where the research strategy is a case study with interviews as a data collection method. The study is based on two perspectives - students and employees. The purpose of the study is to investigate how camera surveillance and management of the data collected is handled at a university, with Uppsala University as a case study. (The focus in the study is on awareness of and attitudes towards camera surveillance - to identify and gain a deeper understanding of university employees and students' attitudes towards camera surveillance and the management of the data that the university collects). The reason for camera surveillance at the University, perceived advantages and disadvantages as well as how the experience and importance of personal integrity, are affected by various factors such as the handling and use of surveillance data. Another issue is the respondents' knowledge of GDPR and the rights they have regarding camera surveillance. The survey is limited to people at Uppsala University. Other monitoring systems such as access systems and logging as well as the handling of this data are delimited from the study and excluded from the survey. / Uppsatsen avser undersöka studenters och universitetsanställdas inställning till kameraövervakning och hanteringen av de data som kan samlas in. Det görs genom en kvalitativ forskningsansats där forskningsstrategin är fallstudie med intervjuer som datainsamlingsmetod. Studien utgår från två perspektiv - studenter och anställda. Syftet med denna studie är att undersöka hur kameraövervakning och hantering av de data som samlas in fungerar på ett universitet, med Uppsala universitet som fallstudie. (Fokus i studien ligger på medvetenhet om och inställning till kameraövervakning - att identifiera och få en djupare förståelse för universitetsanställda och studenters inställning till kameraövervakning och hanteringen av de data som universitetet samlar in). Vad som motiverar (är anledningen till) kameraövervakning på Universitetet, upplevda för- och nackdelar samt hur upplevelsen och vikten av den personliga integriteten, påverkas av olika faktorer som til lexempel hanteringen och användandet av övervakningsdata. En annan frågeställning är respondenternas kunskaper kring GDPR och de rättigheter de har vid kameraövervakning. Undersökningen är avgränsad till personer vid Uppsala universitet. Andra övervakningssystem som till exempel inpasseringssystem och loggning liksom hantering av dessa data är avgränsade från studien och exkluderas från undersökningen.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal