Made available in DSpace on 2014-06-11T19:29:49Z (GMT). No. of bitstreams: 0 Previous issue date: 2010-02-25Bitstream added on 2014-06-13T19:59:46Z : No. of bitstreams: 1 kanda_az_me_ilha.pdf: 1642888 bytes, checksum: 49ebcd86d5690e7be50fcdca35a52a48 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / A proposta deste trabalho é aprimorar a performance da técnica de redução de ruído, subtração espectral baseado na relação SNR a Priori, através da implementação de dois novos parâmetros Potência de Articulação e Não-Articulação obtidas a partir de algumas técnicas psicoacústicas. Faz-se um estudo da anatomia do sistema de audição humana e algumas limitações físicas, com o objetivo de entender o princípio básico da técnica ANIQUE, que é um sistema de avaliação objetiva de voz e têm como princípio o modelamento da percepção humana da voz. Através do modelo ANIQUE são extraídas as principais técnicas psicoacústicas para obtenção dos novos parâmetros, Potência de Articulação e Não- Articulação. Procurou-se apresentar de maneira resumida o processo de equacionamento das técnicas de redução de ruído em sinais de voz e das técnicas psicoacústicas. Posteriormente são descritos todos os processos das técnicas utilizadas que foram simuladas utilizando a linguagem de programação do MatLab®, seguido das avaliações objetivas dos sinais processados pelo software PESQ, que é um programa de avaliação objetiva de voz. Os resultados mostram que a implementação das técnicas psicoacústicas foram eficazes para melhorar a performance da técnica subtração espectral baseada na relação SNR a Priori / The purpose of this work is to enhance the performance of noise reduction techniques based on spectral subtraction, which take in account the a priori signal-to-noise (SNR a Priori) to be estimated considering psychoacoustic criteria. in order to understand the basic principle of the ANIQUE, which is a psychoacoustic based technique used to evaluate the quality of speech signals, it was necessary to develop a study of the anatomy of the human hearing and some physical limitations, From the ANIQUE are extracted new parameters namely Articulation and Non-Articulation Powers, used to estimate the SNR_prio. As a result, it was obtained a new spectral based technique which was implemented in the MatLab® environment and evaluated using the objective quality measure for speech signal simulations namely PESQ. The results show that the implementation of psychoacoustic techniques were effective in enhance the performance of the spectral subtraction technique based on SNR a Priori

Submitted by JÚLIO HEBER SILVA (julioheber@yahoo.com.br) on 2017-06-23T19:10:20Z No. of bitstreams: 2 Dissertação - André Reis de Geus - 2016.pdf: 8269330 bytes, checksum: 1345e49235c545021c88a7baf696f5c0 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Cláudia Bueno (claudiamoura18@gmail.com) on 2017-07-07T20:25:28Z (GMT) No. of bitstreams: 2 Dissertação - André Reis de Geus - 2016.pdf: 8269330 bytes, checksum: 1345e49235c545021c88a7baf696f5c0 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-07-07T20:25:28Z (GMT). No. of bitstreams: 2 Dissertação - André Reis de Geus - 2016.pdf: 8269330 bytes, checksum: 1345e49235c545021c88a7baf696f5c0 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2016-03-10 / Fundação de Amparo à Pesquisa do Estado de Goiás - FAPEG / Insects cause significant losses of stored grains in both quantity and quality. In the scenary, it is of paramount importance an early identification of insects in grains to take control measures. Instead of sampling and visual/laboratory analysis of grains, we propose to carry out the insects identification task automatically, using computational methods to perform endoscopic video analysis. The videos are recorded inside of grains warehouses by an endoscopic camera. As the classification process of moving objects in video rely heavily on precise segmentation of moving objets, we propose a new method of background subtraction and compared their results with the main methods of the literature according to a recent review. The main innovation of the background subtractionmethod rely on the binarization process that uses two thresholds: a global and a local threshold. The binarized results are combined by adding details of the object obtained by the local threshold in the result of the global threshold. Experimental results performed through visual analysis of the segmentation results and using a SVM classifier, suggest that the proposed segmentation method produces more accurate results than the state-of-art background subtraction methods. / Insetos causam perdas quantitativas e qualitativas significantesemgrãos armazenados. Neste cenário, é de vital importância uma identificação rápida de insetos em grãos para que sejam tomadas medidas de controle. Ao invés de coletar amostras de grãos para análise visual/laboratorial, é proposta a realização desta tarefa de identificação de formaautomática, usando métodos computacionais para a análise de vídeos endoscópicos. Os vídeos são gravados dentro de armazéns de grãos usando câmera endoscópica. Como o processo de classificação de objetos em movimento em vídeo depende fundamentalmente de uma segmentação de objeto precisa, é proposto um novo método de segmentação por subtração de plano de fundo e comparado seus resultados com os principais métodos da literatura de acordo com um estudo de revisão recente. A principal inovação neste método de subtração de plano de fundo está no processo de binarização que usa dois thresholds: um global e um local. Os resultados binarizados são combinados pela adição de detalhes do objeto obtido pelo threshold local no resultado do threshold global. Resultados experimentais, realizados através de análise visual dos resultados de segmentação e usandoumclassificadorSVMindicamque o método de segmentação proposto produz melhores resultados que métodos do estado da arte atual da literatura de subtração de plano de fundo.

Desde a introdução dos implantes na Odontologia, controvérsias tem havido acerca da propriedade de conectar os implantes a dentes em próteses fixas. Os implantes osseointegrados não apresentam nenhum ligamento periodontal e são mais firmemente ancorados ao osso. A diferença apresentada nesta ancoragem tem levantado preocupações acerca de o implante osseointegrado, por sua firme fixação ao osso, apresentar pouca flexibilidade para dividir as cargas funcionais com os dentes conectados. O presente estudo teve como propósito acompanhar prospectivamente os parâmetros periodontais ? índice gengival, índice de sangramento, profundidade de sondagem e perda de inserção entre o dente pilar de uma prótese dento-implanto-suportada com o dente colateral do mesmo paciente, e também os parâmetros peri-implantares dos implantes que suportam estas próteses - índice gengival; índice de sangramento; profundidade de sondagem e perda de inserção. Para verificar a perda de inserção foi utilizado o recurso da subtração radiográfica da região em que a prótese foi instalada, com tomadas radiográficas efetuadas no momento da instalação da prótese, decorridos seis meses e após 12 meses de uso, com o objetivo de verificar a ocorrência, ou não, de alteração do suporte ósseo peri-implantar. Foram selecionados 12 pacientes com estado de saúde normal, apresentando área posterior de mandíbula edêntula onde estava indicada a reabilitação protética por meio de prótese fixa unindo dente a implante. Estes foram tratados com implante de um estágio de 10 mm padrão da Straumann (ITI), que apresenta um pescoço polido de 2.8mm, onde a porção do implante que permanece intra-óssea apresenta uma superfície tratada, com diâmetro de 4.1mm. O implante foi instalado no espaço onde se localizaria o retentor mais distal da futura prótese parcial fixa, seguindo o protocolo. As radiografias digitais obtidas foram avaliadas num software de subtração radiográfica, Matrox Inspector versão oito, para verificar a alteração da variação de densidade óptica e contraste das radiografias quando superpostas, sendo os valores observados no implante comparados com os critérios de sucessos , amplamente abordados nos estudos verificados. Todos os implantes instalados nos pacientes osseointegraram, não houve registro de alteração óssea ao redor dos mesmos e nos dentes pilares das próteses em avaliações feitas após seis meses e 12 meses. As próteses continuaram clinicamente em excelente estado após 12 meses de avaliação nas análises clínicas e radiográficas. Ainda, de acordo com o teste de Sinais de Postos de ilcoxon, concluiu-se estatisticamente que não houve diferença nos parâmetros peri-implantares de cada paciente durante os períodos analisados (zero, seis meses e 12 meses), bem como quando comparados com os respectivos dentes colaterais. / Since the introduction of implants in Dentistry, controversies have arisen about the possibility of connecting implants to teeth in fixed prostheses. Osseointegrated implants do no present any periodontal ligature and are more firmly anchored to the bone. The difference presented by this type of anchorage has led to some concern about the limited flexibility of the osseointegrated implant to share the functional loads with the connected teeth. This study intended to follow-up prospectively the periodontal parameters bleeding index, probing depth between the pillar tooth of an implant supported prosthesis with one control tooth of the same patient and also the implants that support theses prostheses. That is why subtraction radiography of the region in which the prosthesis was placed was used, with radiographs taken at prosthesis placement, after six months and one year of use to verify if the periimplant osseous support had or had not undergone changes. Twelve patients with normal health conditions were selected who presented a posterior area edentulous jaw where a prosthetic rehabilitation by a fixed prosthesis joining tooth to implant was indicated. Patients were treated with an implant of one stage of 10mm standard Straumann (ITI) with a 2.8 mm polished neck, in which the remaining intraosseous portion of the implant has a treated surface of 4.1 mm diameter. The implant was placed in the space where according to the protocol, the more distal retainer of the future fixed partial prosthesis would be located. Digital radiographs were assessed with a subtraction radiography software Matrox Inspector version eight, to corroborate the change of contrast in the radiographs when superimposed, with values observed in the implant compared with the success criteria , widely mentioned in surveyed studies. All implants placed in patients became osseointegrated. There was no record of bone change around them and in the dental pillars of the prostheses at the one year follow-up. Prostheses continue to be in excellent conditions after the one year follow-up based upon clinical and radiographic analyses. Furthermore, in accordance with the Wilcoxon test of matched pairs and signed rank, it was concluded that there was no statistical difference between each group of each patient during the time period analyzed (baseline and one year) as well as when the pillars were compared to their respective controls.

A colonização bacteriana no sulco periimplantar ocorre muito precocemente, logo após a cirurgia de instalação dos implantes. Com este intuito, o projeto foi proposto a fim de avaliar o efeito da terapia fotodinâmica (PDT) na manutenção da crista óssea alveolar. Em um estudo de boca dividida, 20 pacientes foram reabilitados com implantes Standard Plus (Straumann®), divididos em dois grupos: um grupo controle (GC) e um grupo submetido à terapia fotodinâmica (GPDT), cujo protocolo foi: densidade de energia 120J/cm2, potência de 40mW, associada ao corante azul de metileno a 0,005%. As aplicações foram realizadas no dia da instalação do implante, após 15, 30 e 45 dias. As coletas microbiológicas foram efetuadas no período inicial e 6ª semana, nos dois grupos, utilizando cones de papel estéril. No GPDT, outras duas coletas foram realizadas após aplicação da terapia: no dia da cirurgia e após 45 dias. As unidades formadoras de colônias totais (UFCt) e pigmentadas de preto (UFCpig) foram contadas. Dados clínicos (Índice de Placa- IP e Sangramento a Sondagem- SS modificados por Mombelli) foram obtidos na 6a semana, 6 meses e 1 ano. Análise radiográfica, com radiografias periapicais, foram efetuadas utilizando a técnica da subtração radiográfica. As películas foram digitalizadas com o escaner SprintScan 35 Plus - Polaroid. Com softwares de manipulação de imagens (Adobe Photoshop CS2®- Califórnia, EUA e ImageJ- National Institutes of Health) foram medidas as distâncias do ombro do implante ao primeiro contato implante-osso, nas seguintes fases: inicial; 6a semana; instalação da prótese, 6 meses e 1 ano após instalação protética. Os testes estatísticos aplicados foram Friedman, Wilcoxon e MannWhitney. Quanto à eficiência da PDT, os resultados mostraram que houve uma redução, estatisticamente significante, tanto para UFCt como UFCpig nos dois momentos aplicados. O número de UFCt e UFCpig do pós-cirúrgico até a 6ª semana apresentou aumento, estatisticamente significante, nos dois grupos. No comparativo entre grupos não houve diferença, estatisticamente significante, entre a diferença do número de UFCt e UFCpig do período inicial ate a 6ª semana. O IP e SS foram similares entre os dois grupos. A avaliação radiográfica mostrou que houve uma perda da crista óssea crescente do período inicial até a instalação da prótese, que se manteve mantém estável após um ano de função para os dois grupos. A média da perda óssea, do período inicial até a instalação da prótese, foi de 0,73mm para o GC e de 0,84mm para o GPDT. Durante o período de função protética as médias de perda óssea foram, respectivamente, 0,27mm e 0,19mm para os GC e GPDT, sem diferença estatisticamente significante entre os grupos. Concluiu-se que: a PDT é eficiente para a redução das UFCt e UFCpig, presentes no sulco periimplantar, durante a osteointegração; a PDT não impede a recolonização, nem o crescimento bacteriano ao longo do tempo, exceto as UFCpig do GPDT, onde a quantidade de bactérias durante a osteointegração manteve-se estatisticamente estável; e com relação à manutenção da crista óssea periimplantar, os GC e GPDT tiveram resultados similares, sem diferença estatisticamente significante. / Bacterial colonization of the peri-implant sulcus occurs very early, right after the implant placement surgery. To this purpose, the project was proposed to evaluate the effect of photodynamic therapy (PDT) in the maintenance of {the} alveolar bone crest. In a split-mouth study, 20 patients were rehabilitated with Standard Plus implants (Straumann®) and they were divided into two groups: a control group (CG) and a group submitted to photodynamic therapy (GPDT), whose protocol was: energy density of 120J/cm2 at power of 40 mW, combined with 0,005% methylene blue. The applications were performed on the day of implant placement, after 15, 30 and 45 days. Microbiological samples were collected in the initial period and after 6 weeks on both groups, using sterile paper cones. In GPDT, two other collections were made after the application of the therapy: on the day of surgery and after 45 days. The total colony forming units (UFCt) and pigmented black units (UFCpig) were counted. Clinical data (Plaque Index-IP and Sulcus bleeding index- SS modified by Mombelli) were obtained at the 6th week, 6 months and 1 year. Radiographic analysis with periapical radiographs were made using the radiographic subtraction technique. The films were digitalized with the scanner SprintScan 35 Plus- Polaroid. Using image manipulation softwares (Adobe Photoshop® CS2- California and ImageJ-National Institutes of Health) distances from the implant shoulder to the first bone-implant contact were measured in phases: initial; 6th week; installation of the prosthesis, 6 months and 1 year after prosthetic installation placement. Friedman, Wilcoxon and Mann-Whitney were the statistical tests applied. Concerning the PDT efficacy, the results presented a statistically significant reduction for both UFCt and UFCpig in the two performed moments. The number of UFCt and UFCpig of the immediate postoperative to 6 weeks indicated a statistically significant increase in both groups. Comparing both the two groups, there was no statistically significant difference between the number of UFCt and UFCpig of the initial period to the 6th week. The IP and SS were similar in both groups. Radiographic evaluation indicated that there was an increasing loss of bone crest during the initial period and the installation of the prosthesis, which remained stable after a year of function for both groups. The bone loss average from the initial period until the installation of the prosthesis was 0,73mm for the GC and 0,84mm for the GPDT. During the prosthetic function the bone loss averages were, respectively, 0,27mm and 0,19mm for the GC and for the GPDT, with no statistically significant difference for both groups. The data led to the conclusion that: PDT is effective in reducing UFCt and UFCpig present in the peri-implant groove during the osseointegration; the PDT does not avoid the re-colonization or bacterial growth over time, except for the UFCpig of GPDT, wherein the amount of bacteria during the osseointegration remained statistically stable; and regarding to the maintenance of the peri-implant bone crest, the GC and GPDT obtained had similar results, with no statistically significant difference.

Made available in DSpace on 2014-08-20T14:30:11Z (GMT). No. of bitstreams: 1 Tese _ Elenara_Ferreira_ Oliveira.pdf: 694369 bytes, checksum: 410d5111d12d3fad68258a97ba292b1b (MD5) Previous issue date: 2006-10-16 / This paper assesses clinical/radiographic changes after incomplete carious dentine removal and tooth sealing. 32 teeth with deep caries lesions were studied. The treatment consisted of incomplete excavation, application of Ca (OH)2 layer, sealing for a 6-7 month period, and then restored. Standardized bitewing radiographs were taken after the temporary sealing, and at 6-7, 14-18, and 36-45 month intervals. The digitized images were analyzed: qualitatively in regard to the radiolucent zone (RZ) depth and tertiary dentine (6-7 and 36-45 months) and quantitatively in regard toradiographic density by image subtraction (6-7, 14-18, and 36-45 months). The subtractions were performed in the RZ beneath the restoration and in two control areas (CA). During the 6-7 month period, there was 1 pulp necroses and 1 pulpexposure, and at the 36-45 month interval, 3 fractures and 3 withdrawals. Nine cases showed a decrease in RZ depth in the 6-7 month period and 3 more cases, at the 36- 45 month period. Four cases displayed tertiary dentine. The radiographic density means and standard deviations were: 129.42±5.83, 127.65±4.67, and 126.86±7.03 (CA, P>0.05) and 132.96 ±7.41, 132.90±5.99, and 132.62±8.68 (RZ, P>0.05) for the 1st, 2nd, and 3rd periods. The radiographic density of the CA differed from the RZ (Tukey test, p<0.05). Interference in microbiological metabolism by partial dentine removal and by tooth sealing arrest lesion progression, suggesting that complete dentine caries removal is not essential to controlling caries progression. / Este trabalho avaliou alterações clínicas/radiográficas após a remoção incompleta da dentina cariada e selamento do dente. Trinta e dois dentes com lesões profundas de cárie foram estudados. O tratamento consistiu da escavação incompleta da dentina cariada, aplicação de uma camada de cimento de Ca(OH)2, selamento provisório por um período de 6-7 meses, e restauração. Radiografias interproximais foram feitas logo após o selamento temporário e em intervalos de 6-7, 14-18, e 36-45 meses. As imagens digitalizadas foram analisadas: qualitativamente em relação à profundidade da zona radiolúcida (ZR) e deposição de dentina terciária (6-7 e 36-45 meses) e quantitativamente, em relação à densidade radiográfica, através da subtração de imagem (6-7, 14-18, e 36-45 meses). As subtrações foram realizadas na ZR, abaixo da restauração, e em duas áreas controles (AC). Durante período de 6-7 meses houve uma necrose e uma exposição pulpar, e no intervalo de 36-45 meses, três fraturas e três perdas de pacientes. Nove casos mostraram uma diminuição da profundidade da ZR no período de 6-7 meses e mais três casos no intervalo de 36-45 meses. Em quatro casos foi observada a deposição de dentina terciária. As médias de densidade radiográfica e desvio-padrão foram: 129.42±5.83, 127.65±4.67, e 126.86±7.03 (AC, p>0.05) e 132.96±7.41, 132.90±5.99, e 132.62±8.68 (ZR, p>0.05) para o 1º, 2º e 3º períodos, respectivamente. A densidade radiográfica das AC foi diferente da ZR (teste de Tukey, p<0.05). A interferência no metabolismo microbiano promovida pela remoção parcial da dentina cariada e selamento do dente provocou a paralisação da lesão, sugerindo que a remoção completa da dentina cariada não é essencial para o controle da progressão da cárie.

Research has shown mathematics education is highly dependent on teaching materials. Here mathematics textbooks have a dominant role in Swedish schools. When curriculum determines the content of textbooks, it promotes confidence in both schools and teachers. The purpose of this study is to analyze and compare textbooks using Basil Bernstein’s theoretical concepts of classification, frames and sequencing to investigate how content and structure are connected to the calculation methods of addition and subtraction. The following questions are posed: In what way have the calculation methods of addition and subtraction of the natural and rational numbers been presented in four selected textbooks for grade 4 in Swedish public schools? What are the similarities and differences in content and structure in these teaching materials? The study finds a variety and selection of material presented in the selected textbooks. Consequently, it is important for teachers to take in to consideration the content and structure of a textbook in mathematics to realize its potential in the classroom.

Foreground segmentation is a common first step in tracking and surveillance applications.  The purpose of foreground segmentation is to provide later stages of image processing with an indication of where interesting data can be found.  This thesis is an investigation of how foreground segmentation can be performed in two contexts: as a pre-step to trajectory tracking and as a pre-step in indoor surveillance applications. Three methods are selected and detailed: a single Gaussian method, a Gaussian mixture model method, and a codebook method.  Experiments are then performed on typical input video using the methods.  It is concluded that the Gaussian mixture model produces the output which yields the best trajectories when used as input to the trajectory tracker.  An extension is proposed to the Gaussian mixture model which reduces shadow, improving the performance of foreground segmentation in the surveillance context.

Le suivi d’objets est de plus en plus utilisé dans les applications de vision par ordinateur. Compte tenu des exigences des applications en termes de performance, du temps de traitement, de la consommation d’énergie et de la facilité du déploiement des systèmes de suivi, l’utilisation des architectures embarquées de calcul devient primordiale. Dans cette thèse, nous avons conçu un système de suivi d’objets pouvant fonctionner en temps réel sur une caméra intelligente de faible coût et de faible consommation équipée d’un processeur embarqué ayant une architecture légère en ressources de calcul. Le système a été étendu pour le suivi d’objets dans un réseau de caméras avec des champs de vision non-recouvrant. La chaîne algorithmique est composée d’un étage de détection basé sur la soustraction de fond et d’un étage de suivi utilisant un algorithme probabiliste Gaussian Mixture Probability Hypothesis Density (GMPHD). La méthode de soustraction de fond que nous avons proposée combine le résultat fournie par la méthode Zipfian Sigma-Delta avec l’information du gradient de l’image d’entrée dans le but d’assurer une bonne détection avec une faible complexité. Le résultat de soustraction est traité par un algorithme d’analyse des composantes connectées afin d’extraire les caractéristiques des objets en mouvement. Les caractéristiques constituent les observations d’une version améliorée du filtre GMPHD. En effet, le filtre GMPHD original ne traite pas les occultations se produisant entre les objets. Nous avons donc intégré deux modules dans le filtre GMPHD pour la gestion des occultations. Quand aucune occultation n’est détectée, les caractéristiques de mouvement des objets sont utilisées pour le suivi. Dans le cas d’une occultation, les caractéristiques d’apparence des objets, représentées par des histogrammes en niveau de gris sont sauvegardées et utilisées pour la ré-identification à la fin de l’occultation. Par la suite, la chaîne de suivi développée a été optimisée et implémentée sur une caméra intelligente embarquée composée de la carte Raspberry Pi version 1 et du module caméra RaspiCam. Les résultats obtenus montrent une qualité de suivi proche des méthodes de l’état de l’art et une cadence d’images de 15 − 30 fps sur la caméra intelligente selon la résolution des images. Dans la deuxième partie de la thèse, nous avons conçu un système distribué de suivi multi-objet pour un réseau de caméras avec des champs non recouvrants. Le système prend en considération que chaque caméra exécute un filtre GMPHD. Le système est basé sur une approche probabiliste qui modélise la correspondance entre les objets par une probabilité d’apparence et une probabilité spatio-temporelle. L’apparence d’un objet est représentée par un vecteur de m éléments qui peut être considéré comme un histogramme. La caractéristique spatio-temporelle est représentée par le temps de transition des objets et la probabilité de transition d’un objet d’une région d’entrée-sortie à une autre. Le temps de transition est modélisé par une loi normale dont la moyenne et la variance sont supposées être connues. L’aspect distribué de l’approche proposée assure un suivi avec peu de communication entre les noeuds du réseau. L’approche a été testée en simulation et sa complexité a été analysée. Les résultats obtenus sont prometteurs pour le fonctionnement de l’approche dans un réseau de caméras intelligentes réel. / Multi-object tracking constitutes a major step in several computer vision applications. The requirements of these applications in terms of performance, processing time, energy consumption and the ease of deployment of a visual tracking system, make the use of low power embedded platforms essential. In this thesis, we designed a multi-object tracking system that achieves real time processing on a low cost and a low power embedded smart camera. The tracking pipeline was extended to work in a network of cameras with nonoverlapping field of views. The tracking pipeline is composed of a detection module based on a background subtraction method and on a tracker using the probabilistic Gaussian Mixture Probability Hypothesis Density (GMPHD) filter. The background subtraction, we developed, is a combination of the segmentation resulted from the Zipfian Sigma-Delta method with the gradient of the input image. This combination allows reliable detection with low computing complexity. The output of the background subtraction is processed using a connected components analysis algorithm to extract the features of moving objects. The features are used as input to an improved version of GMPHD filter. Indeed, the original GMPHD do not manage occlusion problems. We integrated two new modules in GMPHD filter to handle occlusions between objects. If there are no occlusions, the motion feature of objects is used for tracking. When an occlusion is detected, the appearance features of the objects are saved to be used for re-identification at the end of the occlusion. The proposed tracking pipeline was optimized and implemented on an embedded smart camera composed of the Raspberry Pi version 1 board and the camera module RaspiCam. The results show that besides the low complexity of the pipeline, the tracking quality of our method is close to the stat of the art methods. A frame rate of 15 − 30 was achieved on the smart camera depending on the image resolution. In the second part of the thesis, we designed a distributed approach for multi-object tracking in a network of non-overlapping cameras. The approach was developed based on the fact that each camera in the network runs a GMPHD filter as a tracker. Our approach is based on a probabilistic formulation that models the correspondences between objects as an appearance probability and space-time probability. The appearance of an object is represented by a vector of m dimension, which can be considered as a histogram. The space-time features are represented by the transition time between two input-output regions in the network and the transition probability from a region to another. Transition time is modeled as a Gaussian distribution with known mean and covariance. The distributed aspect of the proposed approach allows a tracking over the network with few communications between the cameras. Several simulations were performed to validate the approach. The obtained results are promising for the use of this approach in a real network of smart cameras.

>Magister Scientiae - MSc / The SASL project is in the process of developing a machine translation system that can translate fully-fledged phrases between SASL and English in real-time. To-date, several systems have been developed by the project focusing on facial expression, hand shape, hand motion, hand orientation and hand location recognition and estimation. Achmed developed a highly accurate upper body pose recognition and estimation system. The system is capable of recognizing and estimating the location of the arms from a twodimensional video captured from a monocular view at an accuracy of 88%. The system operates at well below real-time speeds. This research aims to investigate the use of optimizations and parallel processing techniques using the CUDA framework on Achmed’s algorithm to achieve real-time upper body pose recognition and estimation. A detailed analysis of Achmed’s algorithm identified potential improvements to the algorithm. Are- implementation of Achmed’s algorithm on the CUDA framework, coupled with these improvements culminated in an enhanced upper body pose recognition and estimation system that operates in real-time with an increased accuracy.

Magister Scientiae - MSc / Research has shown that five fundamental parameters are required to recognize any sign language gesture: hand shape, hand motion, hand location, hand orientation and facial expressions. The South African Sign Language (SASL) research group at the University of the Western Cape (UWC) has created several systems to recognize sign language gestures using single parameters. These systems are, however, limited to a vocabulary size of 20 – 23 signs, beyond which the recognition accuracy is expected to decrease. The first aim of this research is to investigate the use of two parameters – hand motion and hand shape – to recognise a larger vocabulary of SASL gestures at a high accuracy. Also, the majority of related work in the field of sign language gesture recognition using these two parameters makes use of Hidden Markov Models (HMMs) to classify gestures. Hidden Markov Support Vector Machines (HM-SVMs) are a relatively new technique that make use of Support Vector Machines (SVMs) to simulate the functions of HMMs. Research indicates that HM-SVMs may perform better than HMMs in some applications. To our knowledge, they have not been applied to the field of sign language gesture recognition. This research compares the use of these two techniques in the context of SASL gesture recognition. The results indicate that, using two parameters results in a 15% increase in accuracy over the use of a single parameter. Also, it is shown that HM-SVMs are a more accurate technique than HMMs, generally performing better or at least as good as HMMs.