Tez (Yüksek Lisans) - Süleyman Demirel Üniversitesi, Fen Bilimleri Enstitüsü, Elektronik Bilgisayar Eğitimi Anabilim Dalı, 2007. / Bibliyografya var.

A gerência de redes de computadores cresce a, medida que são necessárias redes mais seguras, menos vulneráveis a falhas e para reduzir a sobrecarga de trabalho dos administradores de rede. Gerenciar elementos físicos das redes de computadores tem sido a tônica da gerencia de redes. Porem, gerenciar as aplicações também a muito importante, pois são o ponto de interação do usuário com a rede. Há grupos trabalhando na padronização da gerencia das aplicações, através da definição de Management Information Bases (MIBs) específicas para as aplicações. Recentemente foi publicada a MIB específica para correio eletrônico Internet [KIL 94]. Alem de grupos oficiais, como o IFIP Working Group 6.6, outras pessoas, também a nível de Brasil, tem desenvolvido trabalhos na áreas de gerencia de aplicações [CAR 94]. Este trabalho propõe um paradigma para gerencia de uma das entidades funcionais do Message Handling System (MHS) X.400, o Agente de Transferência de Mensagens (ATM). O MHS X.400 e a aplicação de correio eletrônico para o ambiente OSI. O correio eletrônico a uma das aplicações mais populares na comunidade de redes de computadores e atualmente tem sido cogitado como base para muitas outras aplicações. A definição conceitual do paradigma de gerencia baseou-se na estrutura de gerencia do modelo OSI, mas a implementação efetiva do protótipo utiliza o protocolo de gerencia da arquitetura Internet, o SNMP. As fases de definição e levantamento dos requisitos para gerencia de um Agente de Transferência de Mansagens (ATM) basearam-se nos modelos organizacional, funcional e informacional definidos pela ISO. 0 modelo organizacional define a hierarquia entre as entidades de gerencia, ou seja, a hierarquia entre agentes e gerentes. O modelo funcional descreve os requisitos para gerencia do ATM segundo as áreas funcionais do modelo: gerencia de fa1has, gerencia de configuração, gerencia de contabilidade, gerencia de desempenho e gerencia de segurança. 0 modelo informacional define a MIB, ou seja, os objetos que serão monitorados. A validação do modelo conceitual foi realizada através da implementação do protótipo. No protótipo foram utilizados os ambientes ISO Development Environment (ISODE), PP e SunNet Manager (SNM). A implementação do protótipo, alem de validar os conceitos te6ricos, foi útil para a obtenção de experiência com relação a integração entre os diversos ambientes. / The network management has been grown due to the need of more secure networks and to reduce the network adminstrator work. The network physical elements management has been the most important topic in network management, but application management is also important because it is the interaction point between users and networks. There are groups working on standardization of application management defining Managemet Information Bases (MIBs) specifics for application management. A Management Information Base for Internet eletronic mail [KIL 941 was recently published. Beyond official groups, such as IFIP Working Group 6.6, there are other people investigating the management process of several well-known applications [CAR 94]. This work proposes a paradigm for the management of one funcional entity of Message Handlig System (MHS) X.400, the Message Transfer Agent (MTA). The MHS X.400 is the eletronic mail application for OSI environment. The eletronic mail is one of most popular network application and it has been indicated to be the base for other applications. The conceptual definition of management paradigm was based in OSI management framework, but the actual prototype implementation uses the SNMP protocol from Internet framework. The definition and elicitation of management requirements for MTA management were based on organization, function and information models defined by ISO. The organization model defines the hierarchy between the management entities, i.e., the hierarchy between agents and managers. The organization model describes the MTA management requirements by the functional areas: fault management, configuration managemet, accounting management, performance management and security managent. The information model defines the MIB, i.e., the managed objects. The conceptual model validation was realized throught prototype implementation. For this validation some softwares were used such as ISO Environment Development (ISODE) and PP that are sharewares commonly used in academic environments and SunNet Manager (SNM) that is a comercial software. This implementation prototype was helpful to apply the network management concepts and to get knowhow about real network managemnt softwares utilization and integration.

A gerência de redes de computadores cresce a, medida que são necessárias redes mais seguras, menos vulneráveis a falhas e para reduzir a sobrecarga de trabalho dos administradores de rede. Gerenciar elementos físicos das redes de computadores tem sido a tônica da gerencia de redes. Porem, gerenciar as aplicações também a muito importante, pois são o ponto de interação do usuário com a rede. Há grupos trabalhando na padronização da gerencia das aplicações, através da definição de Management Information Bases (MIBs) específicas para as aplicações. Recentemente foi publicada a MIB específica para correio eletrônico Internet [KIL 94]. Alem de grupos oficiais, como o IFIP Working Group 6.6, outras pessoas, também a nível de Brasil, tem desenvolvido trabalhos na áreas de gerencia de aplicações [CAR 94]. Este trabalho propõe um paradigma para gerencia de uma das entidades funcionais do Message Handling System (MHS) X.400, o Agente de Transferência de Mensagens (ATM). O MHS X.400 e a aplicação de correio eletrônico para o ambiente OSI. O correio eletrônico a uma das aplicações mais populares na comunidade de redes de computadores e atualmente tem sido cogitado como base para muitas outras aplicações. A definição conceitual do paradigma de gerencia baseou-se na estrutura de gerencia do modelo OSI, mas a implementação efetiva do protótipo utiliza o protocolo de gerencia da arquitetura Internet, o SNMP. As fases de definição e levantamento dos requisitos para gerencia de um Agente de Transferência de Mansagens (ATM) basearam-se nos modelos organizacional, funcional e informacional definidos pela ISO. 0 modelo organizacional define a hierarquia entre as entidades de gerencia, ou seja, a hierarquia entre agentes e gerentes. O modelo funcional descreve os requisitos para gerencia do ATM segundo as áreas funcionais do modelo: gerencia de fa1has, gerencia de configuração, gerencia de contabilidade, gerencia de desempenho e gerencia de segurança. 0 modelo informacional define a MIB, ou seja, os objetos que serão monitorados. A validação do modelo conceitual foi realizada através da implementação do protótipo. No protótipo foram utilizados os ambientes ISO Development Environment (ISODE), PP e SunNet Manager (SNM). A implementação do protótipo, alem de validar os conceitos te6ricos, foi útil para a obtenção de experiência com relação a integração entre os diversos ambientes. / The network management has been grown due to the need of more secure networks and to reduce the network adminstrator work. The network physical elements management has been the most important topic in network management, but application management is also important because it is the interaction point between users and networks. There are groups working on standardization of application management defining Managemet Information Bases (MIBs) specifics for application management. A Management Information Base for Internet eletronic mail [KIL 941 was recently published. Beyond official groups, such as IFIP Working Group 6.6, there are other people investigating the management process of several well-known applications [CAR 94]. This work proposes a paradigm for the management of one funcional entity of Message Handlig System (MHS) X.400, the Message Transfer Agent (MTA). The MHS X.400 is the eletronic mail application for OSI environment. The eletronic mail is one of most popular network application and it has been indicated to be the base for other applications. The conceptual definition of management paradigm was based in OSI management framework, but the actual prototype implementation uses the SNMP protocol from Internet framework. The definition and elicitation of management requirements for MTA management were based on organization, function and information models defined by ISO. The organization model defines the hierarchy between the management entities, i.e., the hierarchy between agents and managers. The organization model describes the MTA management requirements by the functional areas: fault management, configuration managemet, accounting management, performance management and security managent. The information model defines the MIB, i.e., the managed objects. The conceptual model validation was realized throught prototype implementation. For this validation some softwares were used such as ISO Environment Development (ISODE) and PP that are sharewares commonly used in academic environments and SunNet Manager (SNM) that is a comercial software. This implementation prototype was helpful to apply the network management concepts and to get knowhow about real network managemnt softwares utilization and integration.

Orientador: Ana Maria Frattini Fileti / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Quimica / Made available in DSpace on 2018-08-13T14:31:54Z (GMT). No. of bitstreams: 1 Lopreato_LuizGuilhermeRoquette_M.pdf: 1170459 bytes, checksum: af98261941e40b4030182bcc609c0fa6 (MD5) Previous issue date: 2009 / Resumo: A aplicação de redes neurais artificiais (RNA) na solução de problemas de equilíbrio líquido-vapor de misturas multicomponentes apresenta resultados consistentes com valores experimentais observados. Sua principal vantagem, frente às abordagens tradicionais, reside na capacidade de modelar o sistema em equilíbrio sem a necessidade da definição prévia de um modelo específico, tarefa muitas vezes árdua. No presente trabalho utilizam-se dados de três sistemas binários para o treinamento de modelo neural com o objetivo de predizer o equilíbrio de sistema ternário em condições isobáricas. O algoritmo de treinamento utilizado foi o de Levenberg-Marquardt com Regularização Bayesiana, implementado por meio do software MATLAB. A estrutura de rede que apresentou os melhores resultados contém as frações molares dos três componentes na fase líquida, como parâmetros de entrada, e as frações molares dos três componentes na fase vapor e a temperatura como parâmetros de saída. O escalonamento dos dados foi realizado entre 0 e 1, sendo utilizada a função de ativação sigmoidal na camada oculta da rede e a função linear na camada de saída. O desempenho dos modelos foi analisado a partir dos coeficientes de correlação, coeficientes regressão linear, resíduos entre os valores experimentais e preditos, além da raiz dos desvios quadráticos médios (RMSD). O principal sistema estudado foi Acetona-Metanol-Água, sendo os conceitos desenvolvidos extrapolados para os sistemas polares Metanol-Etanol-Água, Metanol-2-Propanol-Água e para o sistema apolar Benzeno-Ciclohexano-Hexano. Os resultados obtidos indicam que a predição de equilíbrio através do modelo neural se mostra consistente com os dados experimentais e as predições realizadas via modelo UNIQUAC. Destaca-se que a construção do modelo neural baseia-se no uso de dados de sistemas binários para a predição de sistemas ternários, enquanto que o modelo UNIQUAC utiliza parâmetros regredidos a partir de dados experimentais para o equilíbrio ternário. Dessa forma, os conceitos utilizados nas duas modelagens são bastante distinto sendo, à priori, favoráveis ao modelo termodinâmico, fato não condizente com os resultados obtidos no presente estudo. / Abstract: Modeling multicomponent vapor-liquid equilibrium (VLE) by means of Artificial Neural Networks (ANN) showed to be able to successfully reproduce observed experimental points. The main advantage of this approach, regards to the traditional approaches, resides in the capacity to modeling the equilibrium condition without the necessity of the prior definition of a specific model, which is a task very often strenuous. In the present work data of three binary systems were used to training a neural network model with the objective to predict the equilibrium of a ternary system at isobaric conditions. The training algorithm used was Levenberg-Marquardt with Bayesian Regularization, implemented through the software MATLAB. The best network structure contains the molar fractions of three components in the liquid phase, as input parameters, and the molar fractions of three components at vapor phase besides the temperature as the output parameters. The data normalization was carried out in the interval between 0 and 1. The sigmoid activation function was used in the hidden layer and linear function was used in the calculations of the exit layer. The performance of the models was analyzed from the coefficients of correlation, linear regression coefficients, residues between the experimental and predicted values, besides the root of the quadratic middle diversions (RMSD). The principal system studied was Methanol-Acetone-Water, being the developed concepts extrapolated to the polar systems Ethanol-Methanol-Water, Methanol-2-propanol-Water and to the nonpolar system Benzene-Cyclohexane-Hexane. Results presented indicate that vapor-liquid equilibrium prediction through neural network is consistent with experimental data and UNIQUAC model predictions. Moreover, when comparing the ANN model with the thermodynamic model it is important to notice that the first model uses binary systems to predict ternary system equilibrium, whereas the thermodynamic model uses parameters obtained from experimental ternary data regressions. Thus, the concepts used in both models are quite different, being favorable to the thermodynamic model. Even so the ANN model showed to be very reliable, in the present study. / Mestrado / Sistemas de Processos Quimicos e Informatica / Mestre em Engenharia Química

Intranet - Erwartungen und Entwicklungen JavaOS - ein Betriebssystem fuer Intranets Network Computer - Eine Verkaufsstrategie? Was gibt es schon? Netscape IFC's - Gestalten von graphische Nutzerinterfaces WWW-Sicherheitsaspekte - ein heisses Thema: Sicherheit im Netz Aufbau einer Certification Authority - z.B. IN e.V. Chemnitz Authentisierung und Verteilung von Konfigurationsfiles - am Bsp. URZ

This dissertation addresses the problem of a lack of a formal incident response capability in the South African National Research and Education Network (SA NREN). While investigating alternatives it was found that no clear method exists to solve this problem. Therefore, a second problem is identified: the lack of a definitive method for establishing a Computer Security Incident Response Team (CSIRT) or Computer Emergency Response Team (CERT) in general. Solving the second problem is important as we then have a means of knowing how to start when building a CSIRT. This will set the basis for addressing the initial problem, resulting in a prepared, improved and coordinated response to IT security incidents affecting the SANREN. To commence, the requirements for establishing a CSIRT are identified via a comprehensive literature review. These requirements are categorized into five areas, namely, the basic business requirements followed by the four Ps of the IT Infrastructure Library (ITIL). That is, People, Processes, Product and Partners, adapted to suit the CSIRT context. Through the use of argumentation, the relationships between the areas are uncovered and explored. Thereafter, a Design Science Research-based process is utilised to develop a generic model for establishing a CSIRT. The model is based on the interactions uncovered between the business requirements and the adapted four Ps. These are summarised through two views -- strategic and tactical -- together forming an holistic model for establishing a CSIRT. The model highlights the decisions required for the business requirements, services, team model and staff, policies and processes, tools and technologies, and partners of a CSIRT respectively. Finally, to address the primary objective, the generic model is applied to the SANREN environment. Thus, the second artefact is an instantiation, a specific model, which can be implemented to create a CSIRT for the SA NREN. To produce the specific model, insight into the nature of the SANREN environment was required. The status quo was revealed through the use of a survey and argumentative analysis of the results. The specific decisions in each area required to establish an SA NREN CSIRT are explored throughout the development of the model. The result is a comprehensive framework for implementing a CSIRT in the SA NREN, detailing the decisions required in each of the areas. This model additionally acts as a demonstration of the utility of the generic model. The implications of this research are twofold. Firstly, the generic model is useful as a basis for anyone wanting to establish a CSIRT. It helps to ensure that all factors are considered and that no important decisions are neglected, thereby enabling an holistic view. Secondly, the specific model for the SA NREN CSIRT serves as a foundation for implementing the CSIRT going forward. It accelerates the process by addressing the important considerations and highlighting the concerns that must be addressed while establishing the CSIRT.

Orientador: Ana Maria Frattini Fileti / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Química / Made available in DSpace on 2018-08-18T19:16:53Z (GMT). No. of bitstreams: 1 Santos_BrunnoFerreirados_M.pdf: 1390100 bytes, checksum: 6d2b1c02098b1bb57ec404e3263f61a6 (MD5) Previous issue date: 2011 / Resumo: Indústrias de polimerização, em geral, utilizam diferentes sistemas reacionais para obter polímeros com diferentes características. Os processos de polimerização de estireno em batelada podem a apresentar um perfil de temperatura transiente, por causa de efeitos de viscosidade no meio reacional, são chamados de efeitos difusionais. Com a finalidade de controlar a temperatura destes tipos de sistema, engenheiros têm se voltado à busca de modelos que descrevam com fidelidade os processos nãolineares. Controle preditivo baseado em um modelo (MPC) de Redes Neurais Artificiais (RNAs) é uma opção para estes casos. Neste trabalho foram implementadas as equações algébricas e os parâmetros ajustados no treinamento de um modelo neural em uma planilha do Microsoft Excel. A atualização das variáveis do modelo se deu através da comunicação entre o Indusoft Web Studio (IWS) e a planilha via driver DDE (Dynamics Data Exchange). O treinamento do modelo neural aconteceu com algoritmo de Levenberg-Marquardt com regularização bayesiana utilizando o toolbox do MATLAB (comando trainbr). A validação do modelo foi realizada online, para garantir o uso do mesmo. Num primeiro momento, foram estudadas as condições para aplicação de controle fixando a proporção da mistura reacional em 50-50% em volume para monômero e solvente (estireno e tolueno, respectivamente) e a modelagem com RNA mostrou bons resultados. Em outro momento, realizou-se um controle em cascata da temperatura do reator manipulando a potência da resistência (variável manipulada) através da passagem de um fluido pela camisa do reator. Foi utilizado o simulink do MATLAB para fazer a malha de controle em cascata, através de um diagrama de blocos e sua atualização aconteceu via protocolo OPC (OLE for Process Control) de dados do Indusoft. Em paralelo, um modelo fenomenológico validado para processos poliméricos, envolvendo equações e parâmetros relacionados à polimerização de estireno, foi utilizado para estimar valores de conversão, massa molar média numérica (Mn) e massa molar média ponderal (Mw). Observou-se o sucesso da aplicação da estratégia de controle cascata adotada para a planta real, permitindo assim estabelecer uma comparação com as predições do modelo. Para o funcionamento do MPC, foram enviadas, a cada 5 segundos, as informações das variáveis de entrada (temperatura do reator, temperatura do fluido e a variação de potência) para o modelo neural na planilha, calculando automaticamente a temperatura do reator em um instante de amostragem à frente usando-a na função-objetivo do Solver no Excel (minimizar erro quadrático). O valor do Solver foi, então, mandado para o processo através de uma programação em VBA (Visual Basic Applications). Foram analisados os desempenhos dos controladores através do oveshoot, tempo de acomodação, IAE e ITAE. E realizou-se a utilização do controlador Híbrido (controle em cascata e MPC) o que melhorou a performance do sistema / Abstract: Chemical processes are included in several types of industries and demand precise control to reach the final product with quality. Because of the viscosity effects of reactor bulk (diffusional effects), the batch styrene polymerization processes present transient temperature behavior. Aiming at controlling the temperature of these systems, rigorous first principles models that describe nonlinearities precisely have been developed by engineers. The model predictive control (MPC) based on neural network modeling is one of the most approppriate techniques for these processes. In this work, the algebraic equations of a neural network and its adjusted parameters from the training step were implemented in an electronic worksheet (Excel software). The communication from the supervisory software (Indusoft Web Studio - IWS) and the worksheet were done through driver DDE (Dynamics Data Exchange), allowing the update of measured variables. The Bayesian Regularization Levenberg-Marquardt algorithm (trainbr function in the software MATLAB) was employed in network training. The model validation was on line performed in order to prove its suitability to the actual process. Initially, a 50/50 % volume ratio of solvent (Toluene) and monomer (styrene) was used for the control application studies, obtaining precise results using the neural network. Afterwards, a temperature cascade control loop was implemented by manipulating the resistance power inverter of the heating thermal fluid tank. This fluid circulates at the reactor jacket. The cascade control diagram was developed in the Simulink/MATLAB software. Updating measurements occurred by means of the OLE for Process Control protocol (OPC). A validated first principles model, using parameters related to the polymerization of styrene, was used to estimate conversion rates, number average molecular weight (Mn) and weight average molecular weight (Mw). The successful application of the control strategy allowed the comparison to the rigorous model predictions. The neural model input variables (reactor temperature, thermal fluid temperature and power inverter position) were captured from sensors every 5 seconds. At this time, the model in the worksheet calculated automatically the one-step-ahead temperature value, using this prediction in the objective function (squared error minimization) implemented in the solver of the Excel software. Using VBA (Visual Basic Applications) programming, the manipulated variable action was calculated and then sent to the process. The overshoot, settling time, IAE and ITAE were used as controller performance indexes. A hybrid controller (cascade feedback and MPC) was also tested, improving the performance of the system / Mestrado / Sistemas de Processos Quimicos e Informatica / Mestre em Engenharia Química

Lenka Šimáková Způsoby a motivace užívání sociálních sítí Facebook a Nyx a identita jejich uživatelů: Kvalitativní komparativní studie Mediální studia, IKSŽ, FSV UK, 2012 Abstract The key goal of this thesis is to find correlations between motivation and usage of social network(ing) sites and how they influence self-presentation of the users as a process of defining their own identity. A qualitative method of semi-structured interviews with five active users of both researched sites was used to describe the impact of the social and technological environment on the way they present themselves towards others. I therefore focus mainly on the social identity of the users: why and how they interact, how they perceive the sites and their role within their groups and how it influences their administration of their personal profiles. I also focus on other relevant subjects of within-line social communication: I describe various aspects of computer mediated communication, user perception of public and private sphere on-line and other issues such as group dynamics or users' approach to the social network(ing) sites as tools for managing their social capital. The resulting set of hypothesis is based on my analysis of the conducted interviews. This thesis brings an insight into user reflection of researched platforms,...

Thesis (MEng)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Within a complex electronic system, sub-system communication forms the backbone of the functionality of any satellite. It allows multiple processors to run simultaneously and data to be shared amongst them. Without it, a single processor would have to control the entire satellite. Not only would such a design then be overly complicated, but the processor would also not have sufficient capacity to service all the components efficiently. Furthermore the detrimental effects that radiation have on integrated circuits are well documented and can be anything from a single bit flip to a complete integrated circuit failure. If not repaired, a failure on a sub-system communication bus could lead to the loss of the entire satellite. Die goal is to create more radiation resistant Controller-Area-Network (CAN) node. Since a full triple modular redundant design will have a large footprint and high power consumption, a combination of techniques will be applied and tested. The goal is to achieve improved footprint utilisation over triple modular redundancy, while still maintaining good resistance to Single Event Upsets (SEU). By applying simulation, it was sufficiently proven that the implementation of the individual techniques used functioned according to expectations. These techniques included error detection and correction using Hamming Codes, single event transient filter and triple modular redundancy. Having applied these mitigation techniques, the footprint of the CAN controller increased by only 116%. Simulation showed that the Error Detection and Correction and Triple Modular Redundancy worked effectively with the CAN controller, and that the CAN controller could function as originally intended. Using radiation testing, the design proved to be more resistant to SEUs than the unmitigated CAN controller. It was thus shown that through using a combination of mitigation techniques, it is possible to develop an optimal design with a high level of resistance against Single Event Upsets, utilizing a smaller footprint than implementing Triple Modular Redundancy. / AFRIKAANSE OPSOMMING: Sub-stelsel kommunikasie vorm die basis van die funksionaliteit in ’n komplekse elektroniese stelsel soos ’n satelliet. Dit skep die vermoë om veelvoudige verwerkers gelyktydig te laat funksioneer en inligting tussen hulle te deel. Sonder sub-stelsel kommunikasie, sal ’n enkele verwerker die hele sateliet moet beheer. Dit sal nie net die hele ontwerp oorkompliseer nie, maar die verwerker sal ook nie genoeg kapisteit hê om al die komponente effektief te diens nie. Die newe-effekte van bestraling op geïntegreerde stroombane is goed gedokumenteer en kan wissel van ’n enkele omgekeerde bis, tot die vernietiging van die geïntegreerde stroombaan. Indien die fout in die kommunikasiestelsel nie herstel word nie, kan dit lei tot die verlies van die hele sateliet. Die doel is om ’n meer bestraling bestande Controller-Area-Network (CAN) nodus te skep. Aangesien ’n volle drie-dubbele-modulêre-oortollige ontwerp ’n baie groot area beslaan en hoë krag verbruik het, gaan ’n kombinasie van versagting tegnieke toegepas en ge-evalueer word. Die doel is om beter area benutting as die drie-dubble-modulêre-oortollige ontwerp te kry, terwyl ’n goeie weerstand teen foute behoue bly. Deur middel van simulasies is voldoende bewyse gelewer dat die implimentasie van die individuele versagting tegnieke soos verwag funktioneer. Hierdie tegnieke sluit in, fout opsporing en regstelling deur middel van Hamming kodes, enkele geval oorgangs verskynsel filter asook drie-dubbele-modulêre-oortollige ontwerp. Nadat versagting meganismes toegepas is, het die area verbruik van die CAN beheerder toegeneem met slegs 116%. Simulasies het bewys dat Fout Opsporing en Regstelling en Drie-Dubbele-Modulêre-Oortollige ontwerp tegnieke binne die CAN beheerder korrek funktioneer, terwyl die CAN beheerder self funktioneer soos dit oorspronklik gefunksioneer het. Deur middel van bestralingstoetse, is dit bewys dat die ontwerp meer bestand is teen foute geïnduseer deur bestraling as die onbeskermde CAN beheerder. Dit is dus bewys dat deur gebruik te maak van verskeie versagting tegnieke dit moontlik is om ’n optimale ontwerp te implimenteer, met ’n hoë weerstand teen foute, maar met ’n laer area verbruik as die van ’n Drie-dubbele-Modulêre-Oortollige ontwerp.

L’estimation de la pose du corps humain est un problème difficile en vision par ordinateur et les actions de toutes les difficultés de détection d’objet. Cette thèse se concentre sur les problèmes de l’estimation de la pose du corps humain dans les images ou vidéo, y compris la diversité des apparences, les changements de scène et l’éclairage de fond de confusion encombrement. Pour résoudre ces problèmes, nous construisons un modèle robuste comprenant les éléments suivants. Tout d’abord, les méthodes top-down et bottom-up sont combinés à l’estimation pose humaine. Nous étendons le modèle structure picturale (PS) de coopérer avec filtre à particules recuit (APF) pour robuste multi-vues estimation de la pose. Deuxièmement, nous proposons plusieurs parties de mélange à base (MMP) modèle d’une partie supérieure du corps pour l’estimation de la pose qui contient deux étapes. Dans la phase de pré-estimation, il y a trois étapes: la détection du haut du corps, catégorie estimation du modèle pour le haut du corps, et la sélection de modèle complet pour pose estimation. Dans l’étape de l’estimation, nous abordons le problème d’une variété de poses et les activités humaines. Enfin, le réseau de neurones à convolution (CNN) est introduit pour l’estimation de la pose. Un Local Multi-résolution réseau de neurones à convolution (LMR-CNN) est proposé pour apprendre la représentation pour chaque partie du corps. En outre, un modèle hiérarchique sur la base LMR-CNN est défini pour faire face à la complexité structurelle des parties de branche. Les résultats expérimentaux démontrent l’efficacité du modèle proposé / Human pose estimation is a challenging problem in computer vision and shares all the difficulties of object detection. This thesis focuses on the problems of human pose estimation in still images or video, including the diversity of appearances, changes in scene illumination and confounding background clutter. To tackle these problems, we build a robust model consisting of the following components. First, the top-down and bottom-up methods are combined to estimation human pose. We extend the Pictorial Structure (PS) model to cooperate with annealed particle filter (APF) for robust multi-view pose estimation. Second, we propose an upper body based multiple mixture parts (MMP) model for human pose estimation that contains two stages. In the pre-estimation stage, there are three steps: upper body detection, model category estimation for upper body, and full model selection for pose estimation. In the estimation stage, we address the problem of a variety of human poses and activities. Finally, a Deep Convolutional Neural Network (DCNN) is introduced for human pose estimation. A Local Multi-Resolution Convolutional Neural Network (LMR-CNN) is proposed to learn the representation for each body part. Moreover, a LMR-CNN based hierarchical model is defined to meet the structural complexity of limb parts. The experimental results demonstrate the effectiveness of the proposed model