Thesis (Ph. D.)--University of Missouri-Columbia, 1996. / Typescript. Vita. Includes bibliographical references (leaves 187-202). Also available on the Internet.

Thesis (M. Eng.), Memorial University of Newfoundland, 1999. / Bibliography: p. 117-119.

Στην παρούσα διπλωματική εργασία παρουσιάζεται μια νέα αρχιτεκτονική δικτύων που ονομάζεται DTN,για δίκτυα τα οποία αντιμετωπίζουν δυσκολίες στην επικοινωνία και στη συνδεσιμότητα των κόμβων τους. Τα δίκτυα αυτά αντιμετωπίζουν τέτοιες δυσκολίες λόγω του δυσμενoύς περιβάλλοντος στο οποίο βρίσκονται,όπως για παράδειγμα ένα δίκτυο στο διάστημα. Έτσι προχωρήσαμε στο σχεδιασμό ενός δικού μας DTN δικτύου που ονομάζεται planet-ece επιλέγοντας την DTN2 υλοποίηση του πρωτοκόλλoυ bundle για τους κόμβους μας καθώς είναι ιδανικότερη για όποιον θέλει να πειραματιστεί. Η ανάπτυξη αυτού του δικτύου έγινε με σκοπό να ερευνηθούν θέματα όπως της διαχείρισης ενός τέτοιου δικτύου (network management) και της ασφάλειας του (security).Βασιζόμενοι τώρα στο εγκατεστημένο πια DTN δίκτυό μας planet-ece και αφού έχει ελεγχθεί η σωστή του λειτουργία, το επόμενο βήμα ήταν να υλοποιηθεί ένας μηχανισμός παρακολούθησης (moniroting), ο οποίος θα μας επιτρέπει να εποπτεύουμε την κατάσταση του δικτύου και της ανταλλασόμενης κίνησης σε επίπεδο DTN. Έπειτα προχωρήσαμε στην υλοποίηση ενός συστήματος παρακολούθησης έξυπνων καμερών μέσω του DTN δικτύου μας planet-ece. Σε αυτή την υλοποίηση λοιπόν, αξιολογούμε την ασφάλεια και τη συμπεριφορά του συστήματος, εστιάζοντας στην επικοινωνία μέσω του δικτύου DTN με σκοπό να εκτιμήσουμε πώς συμπεριφέρεται ένα τέτοιο δίκτυο σε διάφορες πιθανές καταστάσεις, από γεγονότα και δικτυακή διαθεσιμότητα, με ή χωρίς ασφάλεια και σε σύγκριση με υπάρχουσες μεθόδους επικοινωνίας. / This thesis presents a new network architecture called DTN, for networks with difficulties in communication and connectivity. These networks are facing such difficulties due to extreme environmental surroundings, such as a network in space. So we proceeded to design our own DTN network called planet-ece selecting DTN2 implementation of Bundle Protocol for our nodes because is ideal for anyone who wants to experiment. The development of this network was designed to investigate issues such as management of such a network and security. Based now to our installed DTN network named planet-ece and after having verified the correct function, the next step was to implement a monitoring mechanism, which will allow us to monitor the network status and traffic in a DTN level. Then we proceeded to implement a secure DTN-based smart camera surveillance system through our DTN network planet-ece. In this implementation, therefore, we evaluate the safety and behavior of the system, focusing on communication via the DTN network in order to appreciate how such a system behaves in different possible situations, events and network availability, with or without security and in comparison with existing methods of communication.

Made available in DSpace on 2014-06-11T19:30:32Z (GMT). No. of bitstreams: 0 Previous issue date: 2008-11-19Bitstream added on 2014-06-13T19:00:45Z : No. of bitstreams: 1 ruy_rs_dr_prud.pdf: 5800893 bytes, checksum: eca044ea30603eb3cacb3a982ca7088c (MD5) / Nos últimos anos observa-se uma crescente utilização de câmaras digitais em Fotogrametria, especialmente os modelos profissionais de câmaras de pequeno e médio formato. Isso porque os sistemas digitais comerciais de grande formato possuem custos elevados e um complexo sistema de gerenciamento, armazenamento e processamento das imagens. Além disso, as câmaras digitais de pequeno e médio formato possuem algumas vantagens que as tornam altamente atrativas, como: grande disponibilidade no mercado; flexibilidade quanto ao intervalo de focalização; são pequenas, leves e de fácil manejo e; possuem custos substancialmente reduzidos quando comparadas aos sistemas digitais de grande formato. Por outro lado, algumas limitações ainda estão presentes nestes modelos de câmaras, no que se refere à confiabilidade da geometria interna e à resolução dos sensores. Contudo, estudos de caso têm mostrado que estes problemas podem ser contornados, podendo-se utilizar todo o potencial deste tipo de sensor para mapeamentos temáticos, topográficos e cadastrais em áreas de pequeno e médio porte, com grande flexibilidade em relação aos sensores aéreos e orbitais convencionais. Neste contexto, este trabalho teve como objetivo a concepção, implementação física e testes reais de um sistema de aquisição de imagens digitais, formado por câmaras digitais de médio formato integradas a sensores de orientação direta, dispositivos eletrônicos e interfaces de hardware e software. Foram desenvolvidos também estudos, análises, algoritmos e programas computacionais de Fototriangulação com parâmetros adicionais (FPA), com dados de georreferenciamento direto, voltados ao sistema desenvolvido... / In the last years there is a growing use of digital cameras in Photogrammetry, mainly the small and medium format cameras, because of high cost and problems with the images management and postprocessing in the high end digital cameras. Besides, if the small and medium format cameras are calibrated they can provide quality data, together with their advantages: variety in the market; focalization flexibility; are small, light, easy handling and; have low cost if compared with the high resolution cameras. Although, these models of digital cameras have some limitations, like the interior orientation reliability and the resolution of the sensor. Some case studies have showed that these problems can be solved and the digital sensors can be used with success in thematic, topographic and cadastral mapping of small and medium areas, with high flexibility if compared with conventional aerial and orbital sensors. In this context, the aim of this work is the conception, development and real tests performing of a digital image acquisition system composed by medium format digital cameras integrated to direct orientation systems, electronic devices and hardware and software developments. Studies, analysis and computational programs related to block triangulation with additional parameters with direct orientation data were performed for establishing the interior orientation of the cameras that compose the acquisition system... (Complete abstract click electronic access below)

Orientador: Antonio Maria Garcia Tommaselli / Banca: Edson Aparecido Mitishita / Banca: Jorge Luis Nunes e Silva Brito / Banca: Julio Kiyoshi Hasegawa / Banca: Mauricio Galo / Resumo: Nos últimos anos observa-se uma crescente utilização de câmaras digitais em Fotogrametria, especialmente os modelos profissionais de câmaras de pequeno e médio formato. Isso porque os sistemas digitais comerciais de grande formato possuem custos elevados e um complexo sistema de gerenciamento, armazenamento e processamento das imagens. Além disso, as câmaras digitais de pequeno e médio formato possuem algumas vantagens que as tornam altamente atrativas, como: grande disponibilidade no mercado; flexibilidade quanto ao intervalo de focalização; são pequenas, leves e de fácil manejo e; possuem custos substancialmente reduzidos quando comparadas aos sistemas digitais de grande formato. Por outro lado, algumas limitações ainda estão presentes nestes modelos de câmaras, no que se refere à confiabilidade da geometria interna e à resolução dos sensores. Contudo, estudos de caso têm mostrado que estes problemas podem ser contornados, podendo-se utilizar todo o potencial deste tipo de sensor para mapeamentos temáticos, topográficos e cadastrais em áreas de pequeno e médio porte, com grande flexibilidade em relação aos sensores aéreos e orbitais convencionais. Neste contexto, este trabalho teve como objetivo a concepção, implementação física e testes reais de um sistema de aquisição de imagens digitais, formado por câmaras digitais de médio formato integradas a sensores de orientação direta, dispositivos eletrônicos e interfaces de hardware e software. Foram desenvolvidos também estudos, análises, algoritmos e programas computacionais de Fototriangulação com parâmetros adicionais (FPA), com dados de georreferenciamento direto, voltados ao sistema desenvolvido... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: In the last years there is a growing use of digital cameras in Photogrammetry, mainly the small and medium format cameras, because of high cost and problems with the images management and postprocessing in the high end digital cameras. Besides, if the small and medium format cameras are calibrated they can provide quality data, together with their advantages: variety in the market; focalization flexibility; are small, light, easy handling and; have low cost if compared with the high resolution cameras. Although, these models of digital cameras have some limitations, like the interior orientation reliability and the resolution of the sensor. Some case studies have showed that these problems can be solved and the digital sensors can be used with success in thematic, topographic and cadastral mapping of small and medium areas, with high flexibility if compared with conventional aerial and orbital sensors. In this context, the aim of this work is the conception, development and real tests performing of a digital image acquisition system composed by medium format digital cameras integrated to direct orientation systems, electronic devices and hardware and software developments. Studies, analysis and computational programs related to block triangulation with additional parameters with direct orientation data were performed for establishing the interior orientation of the cameras that compose the acquisition system... (Complete abstract click electronic access below) / Doutor

The research involves the analysis and design of a new band of horn antennas, namely Frequency Selective Horns (FSHs). Folding a Frequency Selective Surface (FSS) into a shape of a cone makes a FSH. The stimulus behind this research emanates from an idea to use the FSH as a microwave camera for medical diagnostic purposes. FSHs with dipole geometries have been modelled using existing software. A quasi-static approximation models the dipole elements with a dielectric backing as dielectrically loaded cylindrical conducting elements.

With the growing installation of surveillance video cameras in both private and public areas, it is an immediate requirement to develop intelligent video analysis system for the large-scale camera network. As a prerequisite step of person tracking and person retrieval in intelligent video analysis, person re-identification, which targets in matching person images across camera views is an important topic in computer vision community and has been received increasing attention in the recent years. In the supervised learning methods, the person re-identification task is formulated as a classification problem to extract matched person images/videos (positives) from unmatched person images/videos (negatives). Although the state-of-the-art supervised classification models could achieve encouraging re-identification performance, the assumption that label information is available for all the cameras, is impractical in large-scale camera network. That is because collecting the label information of every training subject from every camera in the large-scale network can be extremely time-consuming and expensive. While the unsupervised learning methods are flexible, their performance is typically weaker than the supervised ones. Though sufficient labels of the training subjects are not available from all the camera views, it is still reasonable to collect sufficient labels from a pair of camera views in the camera network or a few labeled data from each camera pair. Along this direction, we address two scenarios of person re-identification in large-scale camera network in this thesis, i.e. unsupervised domain adaptation and semi-supervised learning and proposed three methods to learn discriminative model using all available label information and domain knowledge in person re-identification. In the unsupervised domain adaptation scenario, we consider data with sufficient labels as the source domain, while data from the camera pair missing label information as the target domain. A novel domain adaptive approach is proposed to estimate the target label information and incorporate the labeled data from source domain with the estimated target label information for discriminative learning. Since the discriminative constraint of Support Vector Machines (SVM) can be relaxed into a necessary condition, which only relies on the mean of positive pairs (positive mean), a suboptimal classification model learning without target positive data can be those using target positive mean. A reliable positive mean estimation is given by using both the labeled data from the source domain and potential positive data selected from the unlabeled data in the target domain. An Adaptive Ranking Support Vector Machines (AdaRSVM) method is also proposed to improve the discriminability of the suboptimal mean based SVM model using source labeled data. Experimental results demonstrate the effectiveness of the proposed method. Different from the AdaRSVM method that using source labeled data, we can also improve the above mean based method by adapting it onto target unlabeled data. In more general situation, we improve a pre-learned classifier by adapting it onto target unlabeled data, where the pre-learned classifier can be domain adaptive or learned from only source labeled data. Since it is difficult to estimate positives from the imbalanced target unlabeled data, we propose to alternatively estimate positive neighbors which refer to data close to any true target positive. An optimization problem for positive neighbor estimation from unlabeled data is derived and solved by aligning the cross-person score distributions together with optimizing for multiple graphs based label propagation. To utilize the positive neighbors to learn discriminative classification model, a reliable multiple region metric learning method is proposed to learn a target adaptive metric using regularized affine hulls of positive neighbors as positive regions. Experimental results demonstrate the effectiveness of the proposed method. In the semi-supervised learning scenario, we propose a discriminative feature learning using all available information from the surveillance videos. To enrich the labeled data from target camera pair, image sequences (videos) of the tagged persons are collected from the surveillance videos by human tracking. To extract the discriminative and adaptable video feature representation, we propose to model the intra-view variations by a video variation dictionary and a video level adaptable feature by multiple sources domain adaptation and an adaptability-discriminability fusion. First, a novel video variation dictionary learning is proposed to model the large intra-view variations and solved as a constrained sparse dictionary learning problem. Second, a frame level adaptable feature is generated by multiple sources domain adaptation using the variation modeling. By mining the discriminative information of the frames from the reconstruction error of the variation dictionary, an adaptability-discriminability (AD) fusion is proposed to generate the video level adaptable feature. Experimental results demonstrate the effectiveness of the proposed method.

In Underwater Acoustic Sensor Networks (UWASNs), cameras have recently been deployed for enhanced monitoring. However, their use has faced several obstacles. Since video capturing and processing consume significant amounts of camera battery power, they are kept in sleep mode and activated only when ultrasonic sensors detect a target. The present study proposes a camera relocation structure in UWASNs to maximize the coverage of detected targets with the least possible vertical camera movement. This approach determines the coverage of each acoustic sensor in advance by getting the most applicable cameras in terms of orientation and frustum of camera in 3-D that are covered by such sensors. Whenever a target is exposed, this information is then used and shared with other sensors that detected the same target. Compared to a flooding-based approach, experiment results indicate that this proposed solution can quickly capture the detected targets with the least camera movement.

Mink (Mustela vison) are an important species because they occur at a high trophic level, they are considered a potential indicator species for environmental contaminants, and they are a popular target species among fur trappers. Despite the importance of mink, their ecology in North America is not well understood. I sampled 59 of 60 Hudson River tributaries with 2 scent stations randomly established along each tributary for 2 10-day monitoring periods to evaluate presence/absence of mink via remote camera photographs and tracks. When sampling was completed, I quantified microhabitat characteristics at 84 randomly selected scent stations. Statistical tests for microhabitat did not indicate a difference among variables selected for analysis between scent stations where mink visits were detected and those where mink visits were not detected. I also analyzed all scent stations for macrohabitat characteristics related to human disturbance, using 150 m circular buffers in ArcGIS 9.2. Although the percent cover of human disturbance was 7.0% greater at scent stations where mink were not detected, the difference was not significant and may therefore indicate that mink may have a tolerance for moderate human disturbances. I also evaluated the performances of 2 types of remote cameras (Moultrie Game Cameras: MGC I–40 and MGC 200) in detecting mink visits at scent stations as well as comparing remote camera detections of mink visits with observations of tracks. The MGC I–40 cameras detected a significantly greater number of mink visits (n = 50) compared to the MGC 200 cameras (n = 3). Detection of mink at scent stations was also significantly greater using the remote cameras compared with observations of mink tracks. Mink were also live–trapped and implanted with subcutaneous radiotransmitters. There were 13 mink captures (0.31 mink/100 trap–nights) with radiotransmitters being implanted in 12 (11 males and 1 female) mink. Overall 166 den sites were located with a mean of 15.9 den sites/mink. Microhabitat analysis of 33 used den sites and 33 unused potentially available den sites (UPADs) indicated that shoreline cover was significantly greater at used den sites. Macrohabitat analysis using 150 m circular buffers surrounding 76 used den sites and 76 unused potentially available sites (UPASs) indicated that human disturbance was 3.3% greater at the UPASs than at used den sites, but this difference was not significant. This may further suggest that mink may have a tolerance for moderate human disturbances. Den site structures most often used by mink included brushpiles or logjams (21.1%) and bank burrows (17.5%). The use of den site structures appeared to be largely based on shoreline cover and availability. The mean linear home range of male mink was 6.6 km (SE = 0.6, range 3.2 – 8.4 km), which was similar to the female's home range of 6.5 km. Ten mink home ranges encompassed portions of both the Hudson River and its tributaries accounting for a mean of 2.8 and 4.4 km, respectively. Mean daily movement distances of mink along shorelines was 659 m (SE = 42, range 0 – 3,087 m) and was significantly greater along tributaries than along the Hudson River.

Dissertação (mestrado) - Universidade Federal de Santa Catarina, Centro Tecnológico, Programa de Pós-graduação em Engenharia Mecânica, Florianópolis, 2103 / Made available in DSpace on 2013-12-06T00:13:42Z (GMT). No. of bitstreams: 1 318757.pdf: 3181400 bytes, checksum: ffc9f7d4673cc588daf5d3a06a66f7d7 (MD5) Previous issue date: 2013 / O objetivo desta dissertação é desenvolver um sistema óptico de medição de baixo custo, para medir superfícies livres. São exemplos de superfícies livres as partes curvilíneas do painel de um carro, o casco de uma lancha ou algo mais simples como embalagens de plástico. A medição dessas superfícies livres, bem como sua comparação é uma tarefa difícil, pois não são facilmente representadas por geometrias regulares, dificultando seu controle geométrico. A medição de superfícies livres, por meio de sistemas ópticos, vem se desenvolvendo continuamente, sendo frequente encontrar equipamentos de medição de alta qualidade; porém, geralmente, com elevados preços. As medições sem contato, que são as medições por meios ópticos, possuem três qualidades importantes que são: incertezas comparáveis a medições por contato, sistemas compactos e a possibilidade de se obter, em pouco tempo, nuvens de pontos contendo milhões de pontos medidos. O presente trabalho utiliza este conhecimento e propõe o desenvolvimento de um sistema de medição óptico de baixo custo com múltiplas câmeras, bem como do algoritmo de funcionamento do sistema elaborado. Os trabalhos experimentais foram desenvolvidos no Laboratório de Metrologia e Automatização (LABMETRO) do Departamento de Pós Graduação da Engenharia Mecânica da UFSC (PósMEC). A aplicação de duas, quatro e seis câmeras foi avaliada neste trabalho; pois, utilizando múltiplas câmeras, tem-se o intuito de obter uma menor incerteza de medição. Esta proposta de montagem com até seis câmeras torna possível extrair mais informações do objeto medido e diminuir a possibilidade de oclusões da cena. A confiabilidade das medições realizadas pelo sistema proposto e a possível melhoria nas incertezas de medição pelo uso de múltiplas câmeras foram testadas pela medição de objetos de dimensões conhecidas, estabelecendo padrões para comparação. A comparação entre medições realizadas pelo sistema óptico de baixo custo e as realizadas com equipamentos de referência possibilitaram avaliar o desempenho do sistema de medição proposto, mostrando a aplicabilidade desta solução e de novas propostas de pesquisas neste campo de estudo <br> / Abstract: The main objective of this dissertation is to develop a low cost measuring optical system, to measure free form surface. Examples of free form surface are the curved parts of a dashboard from a car, the hull of a boat or something simpler like some types of plastic containers. The measuring and comparison of these free surfaces is a difficult task, because they are not easily represented by simple geometries, hindering their geometric control. The measuring of free form surfaces by means of optical systems has been developing continually, being easy to find high quality measuring devices, however, usually requiring high investments. The non-contact measurements, which are measurements of free form surfaces by optical systems, have three important qualities that are: uncertainties comparable to contact measurements, compact systems, and the possibility of obtaining in short time of period a point clouds containing millions of measured points. This work uses this knowledge and proposes the development of a low cost measuring optical system with multiple cameras, as well as the elaborated algorithm to operate the system. The experiments were developed at the Metrology and Automation Laboratory (Labmetro) of the Mechanical Engineering Pos-Graduation Department of UFSC (PosMEC). The application of two, four or six cameras was assessed in this study, because by using multiple cameras it is attained to obtain a lower measurement uncertainty. This assembly proposal with up to six cameras, gives the possibility to extract more data from the measured object and decrease the possibility of occlusions of the scene. The reliability of the measurements taken by the proposed system and possible improvements to the measuring uncertainties by using multiple cameras was tested by the measuring of objects of known dimensions, setting standards for comparison. The comparison between measurements taken by the low cost optical system and the measurements made with a coordinate measuring machine made possible to evaluate the performance of the proposed measurement system, demonstrating the applicability of this solution and new research proposals in this field.