[pt] Sistemas computadorizados de vigilância de veículos têm despertado grande interesse devido à demanda para automatizar tarefas que atualmente são realizadas por operadores humanos. Porém, para realizar estas tarefas é preciso resolver alguns problemas clássicos de visão computacional como sombras, oclusões e variação de iluminação. Este trabalho propõe algoritmos em tempo real para máquinas de baixo custo com o objetivo de rastrear, classificar e determinar a velocidade de cada veículo de uma rodovia. / [en] Vehicle surveillance computerized systems have grown great interest due to the automatizing duties demand, which recently executed by computer vision like shadows, occlusion and light variation have to be solved. The present work proposes real time algorithms for low cost machines focused on tracking, classifying and determining each vehicle`s speed on a highway.

M.Ing. / There is a growing need to enhance situation awareness in the maritime environment utilizing new and current technologies. There are numerous ways to enhance situation awareness by employing long-range vision detection systems, data fusion techniques, such as combining radar and automatic identification system (AIS) data and data mining techniques that allow for filtering out anomalies. With the proliferation of high-quality video equipment and cheaper and faster computational machines, there is an increasing need for automated video surveillance as the amount of information available to the operator for processing is overwhelming. It is therefore necessary that only crucial information that may negatively impact mission effectiveness be presented to the operator. Whilst performing surveillance one would be interested in monitoring other surface vessels within the sensor coverage. The detection and tracking of small and slow moving targets having low signal-to-noise ratios is of interest in the maritime environment. This is particularly challenging as influences from the natural environment, such as sea states, glint, whitecaps and clutter, on a target is captured during image acquisition and this has adverse effects on the tracking of a target. A grey-scale based target tracking algorithm using the particle filter framework was developed and tested in MATLAB® (R2008a). The main focus of the work is on the use of dynamic models in a particle filtering framework. The dynamic model contributes to the propagation of the particles in a particle filtering framework of the target grey-scale distribution. The dynamic models investigated are the constant velocity model and an acceleration model. The algorithm was tested with real-world image sequences in the maritime environment. The targets were tracked for the duration of the image sequence and the dynamic model that accounted for acceleration yielded better results when analysing the position error between the estimated position and the ground truth data points. A slight improvement in this error makes a significant difference on tracking a target as targets in the maritime environment context are small. The future scope of the work would then include accounting for more features of the target such as edge cues and/or implementing adaptive observation models to improve the accuracy, stability and robustness of the algorithm for real-time applications.

The increasingly widespread use of surveillance as a cornerstone of crime control presents novel challenges to questions about personal autonomy, stigmatization, and the shape of social processes. In New York City, the end of stop-and-frisk policing meant the rise of “omnipresence,” built on a progression of the surveillance infrastructure. For residents of NYCHA public housing developments in New York City, the installation of highly visible surveillance structures provokes questions about the role of surveillance in increasing contact with the criminal justice system, as well about how use of these structures redefines spaces against a background of gentrification, a globalizing real estate market, and unbuffered income stratification. This project uses ethnographic methods, including interviews and participant observation to explore the phenomenology of social processes undertaken by individuals in relation to surveillance structures and to interrogate the use of surveillance in public housing. Engaging with the work of Bourdieu, Lefebvre, Wacquant, Goffman, and Sassen, the dissertation explores social ordering, stigmatization, norm durability, and place-making among NYCHA residents. The experiences of residents of neighborhoods that are on both sides of the city-wide demographic shifts associated with gentrification are further contrasted, contextualizing their interactions at the macro and micro levels. At the policy level, NYCHA housing becomes a focus of crime control measures as if containing public housing will address the root causes of crime. On the ground, however, cameras do not work to prevent crime, even when they are in good repair. Lights may make some individuals feel safer walking through courtyards late at night, but those same individuals fear that danger is just beyond the reach of the lights in the stairwells or playgrounds. If the cameras and lights fail to allay the fears of those in public housing, if their experiences with the system of surveillance have proven unsuccessful in preventing crime, giving the feeling of safety, or helping to solve crime, what is their purpose? This study posits that these structures are a spectacle. They are structures heavily laden with symbolism that reassure non-residents and residents alike of the neutralization of NYCHA residents. This symbolic dynamic ultimately stigmatizes NYCHA residents and pushes them further towards the systemic edge. Among the theoretical implications of the dissertation’s conclusions are an enhanced understanding of the connections between the persistence of social inequality, the “terra non grata” of certain urban spaces, and the dismantling of the social welfare state. The practical implications of this work are significant and add to discourses around the function of technology in the creation of new types of barriers.

Thesis (MSc)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Object detection and tracking are important components of many computer vision applications including automated surveillance. Automated surveillance attempts to solve the challenges associated with closed-circuit camera systems. These include monitoring large numbers of cameras and the associated labour costs, and issues related to targeted surveillance. Object detection is an important step of a surveillance system and must overcome challenges such as changes in object appearance and illumination, dynamic background objects like ickering screens, and shadows. Our system uses Gaussian mixture models, which is a background subtraction method, to detect moving objects. Tracking is challenging because measurements from the object detection stage are not labelled and could be from false targets. We use multiple hypothesis tracking to solve this measurement origin problem. Practical long-term tracking of objects should have re-identi cation capabilities to deal with challenges arising from tracking failure and occlusions. In our system each tracked object is assigned a one-class support vector machine (OCSVM) which learns the appearance of that object. The OCSVM is trained online using HSV colour features. Therefore, objects that were occluded or left the scene can be reidenti ed and their tracks extended. Standard, publicly available data sets are used for testing. The performance of the system is measured against ground truth using the Jaccard similarity index, the track length and the normalized mean square error. We nd that the system performs well. / AFRIKAANSE OPSOMMING: Die opsporing en volging van voorwerpe is belangrike komponente van baie rekenaarvisie toepassings, insluitend outomatiese bewaking. Outomatiese bewaking poog om die uitdagings wat verband hou met geslote kring kamera stelsels op te los. Dit sluit in die monitering van groot hoeveelhede kameras en die gepaardgaande arbeidskoste, en kwessies wat verband hou met toegespitse bewaking. Die opsporing van voorwerpe is 'n belangrike stap in 'n bewakingstelsel en moet uitdagings soos veranderinge in die voorwerp se voorkoms en beligting, dinamiese agtergrondvoorwerpe soos ikkerende skerms, en skaduwees oorkom. Ons stelsel maak gebruik van Gaussiese mengselmodelle, wat 'n agtergrond-aftrek metode is, om bewegende voorwerpe op te spoor. Volging is 'n uitdaging, want afmetings van die voorwerp-opsporing stadium is nie gemerk nie en kan afkomstig wees van valse teikens. Ons gebruik verskeie hipotese volging (multiple hypothesis tracking ) om hierdie meting-oorsprong probleem op te los. Praktiese langtermynvolging van voorwerpe moet heridenti seringsvermoëns besit, om die uitdagings wat voortspruit uit mislukte volging en okklusies te kan hanteer. In ons stelsel word elke gevolgde voorwerp 'n een-klas ondersteuningsvektormasjien (one-class support vector machine, OCSVM) toegeken, wat die voorkoms van daardie voorwerp leer. Die OCSVM word aanlyn afgerig met die gebruik van HSV kleurkenmerke. Daarom kan voorwerpe wat verdwyn later her-identi seer word en hul spore kan verleng word. Standaard, openbaar-beskikbare datastelle word vir toetse gebruik. Die prestasie van die stelsel word gemeet teen korrekte afvoer, met behulp van die Jaccard ooreenkoms-indeks, die spoorlengte en die genormaliseerde gemiddelde kwadraatfout. Ons vind dat die stelsel goed presteer.

Not available / text

The care of individuals with concerns about development, health, and wellness is often a difficult, complicated task and may rely on a team of diverse caregivers. There are many decisions that caregivers must make to help ensure that the best care and health monitoring are administered. For my dissertation work, I have explored the use of embedded capture and access to support decision-making for caregivers. Embedded capture and access integrates simple and unobtrusive capture and useful access, including trending information and rich data, into existing work practices. I hypothesized that this technology encourages more frequent access to evidence, increased collaboration amongst caregivers, and decisions made with higher confidence. I have explored this technology through real world deployments of new embedded capture and access applications in two domains. For the first domain, I have developed two applications to support decision-making for caregivers administering therapy to children with autism. The first application, Abaris, supports therapists working with a single child in a home setting, and the second application, Abaris for Schools, extends the ideas of Abaris for use in a school setting for many teachers working with multiple children. The second domain I have explored is decision-making for parents of newborn children. In particular, I developed and evaluated embedded capture and access technology to support parents, pediatricians, and secondary childcare providers in making decisions about whether a child s development is progressing normally in order to promote the earlier detection of developmental delays.

Cette thèse s’intéresse aux méthodes de classification par Machines à Vecteurs de Support (SVM) partiellement supervisées permettant la détection de nouveauté (One-Class SVM). Celles-ci ont été étudiées dans le but de réaliser la détection d’événements audio anormaux pour la surveillance d’infrastructures publiques, en particulier dans les transports. Dans ce contexte, l’hypothèse « ambiance normale » est relativement bien connue (même si les signaux correspondants peuvent être très non stationnaires). En revanche, tout signal « anormal » doit pouvoir être détecté et, si possible, regroupé avec les signaux de même nature. Ainsi, un système de référence s’appuyant sur une modélisation unique de l’ambiance normale est présenté, puis nous proposons d’utiliser plusieurs SVM de type One Class mis en concurrence. La masse de données à traiter a impliqué l’étude de solveurs adaptés à ces problèmes. Les algorithmes devant fonctionner en temps réel, nous avons également investi le terrain de l’algorithmie pour proposer des solveurs capables de démarrer à chaud. Par l’étude de ces solveurs, nous proposons une formulation unifiée des problèmes à une et deux classes, avec et sans biais. Les approches proposées ont été validées sur un ensemble de signaux réels. Par ailleurs, un démonstrateur intégrant la détection d’événements anormaux pour la surveillance de station de métro en temps réel a également été présenté dans le cadre du projet Européen VANAHEIM / This thesis addresses partially supervised Support Vector Machines for novelty detection (One-Class SVM). These have been studied to design abnormal audio events detection for supervision of public infrastructures, in particular public transportation systems. In this context, the null hypothesis (“normal” audio signals) is relatively well known (even though corresponding signals can be notably non stationary). Conversely, every “abnormal” signal should be detected and, if possible, clustered with similar signals. Thus, a reference system based on a single model of normal signals is presented, then we propose to use several concurrent One-Class SVM to cluster new data. Regarding the amount of data to process, special solvers have been studied. The proposed algorithms must be real time. This is the reason why we have also investigated algorithms with warm start capabilities. By the study of these algorithms, we have proposed a unified framework for One Class and Binary SVMs, with and without bias. The proposed approach has been validated on a database of real signals. The whole process applied to the monitoring of a subway station has been presented during the final review of the European Project VANAHEIM

The purpose of this exploratory study was to examine resident, family member and staff perceptions of electronic monitoring and their effect on resident rights. The sample consisted of 53 nursing home residents, 104 staff and 25 family members, in the Dallas Fort Worth metroplex, from a nursing facility in which residents utilize video cameras in their rooms (Nursing Facility 1), two nursing facilities that have video cameras in their common rooms areas (Nursing Facility 2 and 3) and a nursing facility that does not utilize video cameras (Nursing Facility 4). The interview questions and self-administered surveys were in regard to the participant's perceptions of electronic monitoring, perceived risks and benefits of video cameras, awareness of resident rights and consciousness of potential risks to resident rights. Data were analyzed using a mixed methods approach using both ATLAS t.i and SAS. Study findings revealed that residents, family members and staff are aware of the potential benefits of electronic monitoring in nursing facilities. While respondents are hesitant to have electronic monitoring in resident rooms, they are interested in utilizing electronic monitoring in common areas. While residents and staff believe that electronic monitoring compromises resident rights, family members believe resident rights are protected. Different types of staff have different perceptions of electronic monitoring. Those staff members that are more directly involved in resident care are less accepting of electronic monitoring compared to staff that have episodic visits with residents. Among staff members, nursing facilities with prior experience with electronic monitoring are less accepting of electronic monitoring. Further studies are needed to enhance this research.

Recently, there is a great interest in moving object tracking in the fields of security and surveillance. Object recognition under partial occlusion is the core of any object tracking system. This thesis presents an automatic and real-time color object-recognition system which is not only robust but also occlusion tolerant. The intended use of the system is to recognize and track external vehicles entered inside a secured area like a school campus or any army base. Statistical morphological skeleton is used to represent the visible shape of the vehicle. Simple curve matching and different feature based matching techniques are used to recognize the segmented vehicle. Features of the vehicle are extracted upon entering the secured area. The vehicle is recognized from either a digital video frame or a static digital image when needed. The recognition engine will help the design of a high performance tracking system meant for remote video surveillance.

Indiana University-Purdue University Indianapolis (IUPUI) / Surveillance videos are recorded continually and the retrieval of such videos currently still relies on human operators. Automatic retrieval has not reached a satisfactory accuracy. As an intermediate representation, this work develops multiple original temporal profiles of video to convey accurate temporal information in the video while keeping certain spatial characteristics. These are effective methods to visualizes surveillance video contents efficiently in a 2D temporal image, suitable for indexing and retrieving a large video database. We are aiming to provide a compact index that is intuitive and preserves most of the information in the video in order to avoid browsing extensive video clips frame by frame. By considering some of the properties of static surveillance videos, we aim at accentuating the temporal dimension in our visualization. We have introduced our framework as three unique methods that visualize different aspects of a surveillance video, plus an extension to non-static surveillance videos. In our first method "Localized Temporal Profile", by knowing that most surveillance videos are monitoring specific locations, we try to emphasize the other dimension, time, in our solution. we focus on describing all the events only in critical locations of the video. In our next method "Multi-Position Temporal Profile", we generate an all-inclusive profile that covers all the events in the video field of view. In our last method "Motion Temporal Profile" we perform in-depth analysis of scene motion and try to handle targets with non-uniform, non-translational motion in our temporal profile. We then further extend our framework by loosening the constraint that the video is static and including cameras with smooth panning motion as such videos are widely used in practice. By performing motion analysis on the camera, we stabilize the camera to create a panorama-like effect for the video, allowing us to utilize all of the aforementioned methods. The resulting profiles allows temporal indexing to each video frame, and contains all spatial information in a continuous manner. It also shows the actions and progress of events in the temporal profile. Flexible browsing and effective manipulation of videos can be achieved using the resulting video profiles.