Este projeto de doutorado apresenta uma nova abordagem MAP-MRF para a classificação contextual de imagens multiespectrais utilizando combinação de modelos de Campos Aleatórios Markovianos definidos em sistemas de ordens superiores. A modelagem estatística para o problema de classificação segue o paradigma Bayesiano, com a definição de um modelo Markoviano para os dados observados (Gaussian Markov Random Field multiespectral) e outro modelo para representar o conhecimento a priori (Potts). Nesse cenário, o parâmetro β do modelo de Potts atua como um parâmetro de regularização, tendo papel fundamental no compromisso entre as observações e o conhecimento a priori, de modo que seu correto ajuste é necessário para a obtenção de bons resultados. A introdução de sistemas de vizinhança de ordens superiores requer a definição de novos métodos para a estimação dos parâmetros dos modelos Markovianos. Uma das contribuições desse trabalho é justamente propor novas equações de pseudo-verossimilhança para a estimação desses parâmetros no modelo de Potts em sistemas de segunda e terceira ordens. Apesar da abordagem por máxima pseudo-verossimilhança ser amplamente utilizada e conhecida na literatura de campos aleatórios, pouco se conhece acerca da acurácia dessa estimação. Foram derivadas aproximações para a variância assintótica dos estimadores propostos, caracterizando-os completamente no caso limite, com o intuito de realizar inferências e análises quantitativas sobre os parâmetros dos modelos Markovianos. A partir da definição dos modelos e do conhecimento dos parâmetros, o próximo estágio é a classificação das imagens multiespectrais. A solução para esse problema de inferência Bayesiana é dada pelo critério de estimação MAP, onde a solução ótima é determinada maximizando a probabilidade a posteriori, o que define um problema de otimização. Como não há solução analítica para esse problema no caso de prioris Markovianas, algoritmos iterativos de otimização combinatória foram empregados para aproximar a solução ótima. Nesse trabalho, adotam-se três métodos sub-ótimos: Iterated Conditional Modes, Maximizer of the Posterior Marginals e Game Strategy Approach. Porém, é demonstrado na literatura que tais métodos convergem para máximos locais e não globais, pois são altamente dependentes de sua condição inicial. Isto motivou o desenvolvimento de uma nova abordagem para combinação de classificadores contextuais, que utiliza múltiplas inicializações simultâneas providas por diferentes classificadores estatísticos pontuais. A metodologia proposta define um framework MAP-MRF bastante robusto para solução de problemas inversos, pois permite a utilização e a integração de diferentes condições iniciais em aplicações como classificação, filtragem e restauração de imagens. Como medidas quantitativas de desempenho, são adotados o coeficiente Kappa de Cohen e o coeficiente Tau de Kendall para verificar a concordância entre as saídas dos classificadores e a verdade terrestre (amostras pré-rotuladas). Resultados obtidos mostram que a inclusão de sistemas de vizinhança de ordens superiores é de fato capaz de melhorar significativamente não apenas o desempenho da classificação como também a estimação dos parâmetros dos modelos Markovianos, reduzindo tanto o erro de estimação quanto a variância assintótica. Além disso, a combinação de classificadores contextuais através da utilização de múltiplas inicializações simultâneas melhora significativamente o desempenho da classificação se comparada com a abordagem tradicional com apenas uma inicialização. / This work presents a novel MAP-MRF approach for multispectral image contextual classification by combining higher-order Markov Random Field models. The statistical modeling follows the Bayesian paradigm, with the definition of a multispectral Gaussian Markov Random Field model for the observations and a Potts MRF model to represent the a priori knowledge. In this scenario, the Potts MRF model parameter (β) plays the role of a regularization parameter by controlling the tradeoff between the likelihood and the prior knowledge, in a way that a suitable tunning for this parameter is required for a good performance in contextual classification. The introduction of higher-order MRF models requires the specification of novel parameter estimation methods. One of the contributions of this work is the definition of novel pseudo-likelihood equations for the estimation of these MRF parameters in second and third order neighborhood systems. Despite its widely usage in practical MRF applications, little is known about the accuracy of maximum pseudo-likelihood approach. Approximations for the asymptotic variance of the proposed MPL estimators were derived, completely characterizing their behavior in the limiting case, allowing statistical inference and quantitative analysis. From the statistical modeling and having the model parameters estimated, the next step is the multispectral image classification. The solution for this Bayesian inference problem is given by the MAP criterion, where the optimal solution is obtained by maximizing the a posteriori distribution, defining an optimization problem. As there is no analytical solution for this problem in case of Markovian priors, combinatorial optimization algorithms are required to approximate the optimal solution. In this work, we use three suboptimal methods: Iterated Conditional Modes, Maximizer of the Posterior Marginals and Game Strategy Approach, a variant approach based on non-cooperative game theory. However, it has been shown that these methods converge to local maxima solutions, since they are extremelly dependent on the initial condition. This fact motivated the development of a novel approach for combination of contextual classifiers, by making use of multiple initializations at the same time, where each one of these initial conditions is provided by different pointwise pattern classifiers. The proposed methodology defines a robust MAP-MRF framework for the solution of general inverse problems since it allows the use and integration of several initial conditions in a variety of applications as image classification, denoising and restoration. To evaluate the performance of the classification results, two statistical measures are used to verify the agreement between the classifiers output and the ground truth: Cohens Kappa and Kendalls Tau coefficient. The obtained results show that the use of higher-order neighborhood systems is capable of significantly improve not only the classification performance, but also the MRF parameter estimation by reducing both the estimation error and the asymptotic variance. Additionally, the combination of contextual classifiers through the use of multiple initializations also improves the classificatoin performance, when compared to the traditional single initialization approach.

A ruptura de materiais heterogêneos quase frágeis, como o concreto, as cerâmicas e diferentes tipos de rochas, tem um comportamento mecânico complexo, foco de estudo de pesquisadores já há muito tempo. Uma ferramenta para simular o comportamento até ruptura destes tipos de materiais são os métodos de elementos discretos formados por barras (Discrete Element Method - DEM). Neste método as massas são concentradas em pontos nodais que estão ligados por meio de elementos unidimensionais caracterizados por relações constitutivas uniaxiais simples. Neste contexto, realizam-se novas implementações na versão do DEM com o intuito de estudar diversos aspectos dos materiais quase frágeis. Os principais focos da tese estão na desvinculação do nível de discretização do modelo do comprimento de correlação dos campos aleatórios que caracterizam as propriedades mecânicas. Através da interpretação dos resultados na simulação de emissão acústica é possível conhecer melhor o processo de dano neste tipo de material. Testes experimentais preliminares também são apresentados. Também são realizadas simulações em DEM utilizando carregamento biaxial, onde é explorado nessas simulações o comportamento dos modelos sob a ótica da técnica de emissão acústica. / The rupture of heterogeneous materials quasi-fragile, like concrete, ceramics and different types of rocks have a complex mechanical behavior, which has been the focus of study by researchers for a long time. The truss-like Discrete Element Method (DEM) was used to perform numerical simulations of the testing processes. The test results and the results of the numerical analyses, in terms of load vs. time diagram and AE data, as determined through b-value. In this context, new implementations in the version of DEM were realized with the objective to study the aspects of quasi-fragile materials. The main focuses of the thesis are in untying the level of discretization of the model of the correlation length from random fields which characterize the mechanical properties. And, through the interpretation of the results in the simulation of the acoustic emission, it is possible to better understand the process of damage in this type of material. Are also performed in DEM the biaxial loading, where is explored in this simulations the behavior of models from the point of view of the acoustic emission technique.

In this thesis we propose to use a penalized maximum conditional likelihood criterion to estimate the graph of a general discrete Markov random field. We prove the almost sure convergence of the estimator of the graph in the case of a finite or countable infinite set of variables. Our method requires minimal assumptions on the probability distribution and contrary to other approaches in the literature, the usual positivity condition is not needed. We present several examples with a finite set of vertices and study the performance of the estimator on simulated data from theses examples. We also introduce an empirical procedure based on k-fold cross validation to select the best value of the constant in the estimators definition and show the application of this method in two real datasets. / Nesta tese propomos um critério de máxima verossimilhança penalizada para estimar o grafo de dependência condicional de um campo aleatório Markoviano discreto. Provamos a convergência quase certa do estimador do grafo no caso de um conjunto finito ou infinito enumerável de variáveis. Nosso método requer condições mínimas na distribuição de probabilidade e contrariamente a outras abordagens da literatura, a condição usual de positividade não é necessária. Introduzimos alguns exemplos com um conjunto finito de vértices e estudamos o desempenho do estimador em dados simulados desses exemplos. Também propomos um procedimento empírico baseado no método de validação cruzada para selecionar o melhor valor da constante na definição do estimador, e mostramos a aplicação deste procedimento em dois conjuntos de dados reais.

Estruturas do tipo linha/curva (line-like, curve-like), alongadas e ramificadas são comumente encontradas nos ecossistemas que conhecemos. Na biomedicina e na biociências, por exemplo, diversas aplicações podem ser observadas. Justamente por este motivo, extrair este tipo de estrutura em imagens é um constante desafio em problemas de análise de imagens. Porém, diversas dificuldades estão envolvidas neste processo. Normalmente as características espectrais e espaciais destas estruturas podem ser muito complexas e variáveis. Especificamente as mais \"finas\" são muito frágeis a qualquer tipo de processamento realizado na imagem e torna-se muito fácil a perda de informações importantes. Outro problema bastante comum é a ausência de parte das estruturas, seja por motivo de pouca resolução, ou por problemas de aquisição, ou por casos de oclusão. Este trabalho tem por objetivo explorar, descrever e desenvolver técnicas de detecção/segmentação de estruturas finas e ramificadas. Diferentes métodos são utilizados de forma combinada, buscando uma melhor representação topológica e perceptual das estruturas e, assim, melhores resultados. Grafos são usados para a representação das estruturas. Esta estrutura de dados vem sendo utilizada com sucesso na literatura na resolução de diversos problemas em processamento e análise de imagens. Devido à fragilidade do tipo de estrutura explorado, além das técnicas de processamento de imagens, princípios de visão computacional são usados. Busca-se, desta forma, obter um melhor \"entendimento perceptual\" destas estruturas na imagem. Esta informação perceptual e informações contextuais das estruturas são utilizadas em um modelo de campos aleatórios de Markov, buscando o resultado final da detecção através de um processo de otimização. Finalmente, também propomos o uso combinado de diferentes modalidades de imagens simultaneamente. Um software é resultado da implementação do arcabouço desenvolvido e o mesmo é utilizado em duas aplicações para avaliar a abordagem proposta: extração de estradas em imagens de satélite e extração de raízes em imagens de perfis de solo. Resultados do uso da abordagem proposta na extração de estradas em imagens de satélite mostram um melhor desempenho em comparação com método existente na literatura. Além disso, a técnica de fusão proposta apresenta melhora significativa de acordo com os resultados apresentados. Resultados inéditos e promissores são apresentados na extração de raízes de plantas. / Line- curve-like, elongated and ramified structures are commonly found inside many known ecosystems. In biomedicine and biosciences, for instance, different applications can be observed. Therefore, the process to extract this kind of structure is a constant challenge in image analysus problems. However, various difficulties are involved in this process. Their spectral and spatial characteristics are usually very complex and variable. Considering specifically the thinner ones, they are very \"fragile\" to any kind of process applied to the image, and then, it becomes easy the loss of crucial data. Another very common problem is the absence of part of the structures, either because of low image resolution and image acquisition problems or because of occlusion problems. This work aims to explore, describe and develop techniques for detection/segmentation of thin and ramified structures. Different methods are used in a combined way, aiming to reach a better topological and perceptual representation of the structures and, therefore, better results. Graphs are used to represent the structures. This data structure has been successfully used in the literature for the development of solutions for many image processing and analysis problems. Because of the fragility of the kind of structures we are dealing with, some computer vision principles are used besides usual image processing techniques. In doing so, we search for a better \"perceptual understanding\" of these structures in the image. This perceptual information along with contextual information about the structures are used in a Markov random field, searching for a final detection through an optimization process. Lastly, we propose the combined use of different image modalities simultaneously. A software is produced from the implementation of the developed framework and it is used in two application in order to evaluate the proposed approach: extraction of road networks from satellite images and extraction of plant roots from soil profile images. Results using the proposed approach for the extraction of road networks show a better performance if compared with an existent method from the literature. Besides that, the proposed fusion technique presents a meaningful improvement according to the presented results. Original and promising results are presented for the extraction of plant roots from soil profile images.

When moving into a more connected world together with machines, a mutual understanding will be very important. With the increased availability in wear- able sensors, a better understanding of human needs is suggested. The Dart- mouth Research study at the Psychiatric Research Center has examined the viability of detecting and further on predicting human behaviour and complex tasks. The field of smoking detection was challenged by using the Q-sensor by Affectiva as a prototype. Further more, this study implemented a framework for future research on the basis for developing a low cost, connected, device with Thayer Engineering School at Dartmouth College. With 3 days of data from 10 subjects smoking sessions was detected with just under 90% accuracy using the Conditional Random Field algorithm. However, predicting smoking with Electrodermal Momentary Assessment (EMA) remains an unanswered ques- tion. Hopefully a tool has been provided as a platform for better understanding of habits and behaviour.

Les ouvrages hydrauliques – barrages et digues – sont des ouvrages de génie civil à risque. Leur rupture engendre des conséquences humaines et matérielles souvent dramatiques. Parmi eux, les barrages en remblai représentent une part importante du parc de barrages au niveau national comme mondial, auxquels s’ajoutent d’importants linéaires de digues en remblai fluviales et maritimes. La sécurité structurale de ces ouvrages est traditionnellement évaluée par des méthodes déterministes ou semi-probabilistes aux états-limites. Cependant, la réglementation française en matière d’ouvrages hydrauliques a récemment évolué en préconisant pour les grands barrages la réalisation d’études de dangers (EDD) basées sur les méthodes d’analyse de risques et impliquant l’utilisation de démarches probabilistes. Dans ce cadre, l’objectif principal de la thèse est de développer une démarche probabiliste pour l’évaluation de la fiabilité structurale des ouvrages hydrauliques en remblai vis-à-vis du mécanisme de glissement, qui constitue l’état-limite conditionnant la géométrie de ces ouvrages. Le développement d’une telle démarche nécessite de traiter trois questions scientifiques principales : · l’élaboration d’un modèle hydromécanique pour l’évaluation déterministe de la stabilité de l’ouvrage vis-à-vis du mécanisme de glissement ; · la modélisation probabiliste de la variabilité spatiale des propriétés mécaniques et hydrauliques des matériaux constituant le remblai ; · le couplage mécano-fiabiliste intégrant les modèles de variabilité spatiale au modèle hydromécanique. De nombreux travaux de recherche ont été réalisés sur ces questions et sont disponibles dans la littérature scientifique. Cependant, ils ne traitent qu’une partie des aspects de la problématique globale d’évaluation de la fiabilité et l’absence de recherches intégratrices est à déplorer. Notre travail, propose une démarche méthodologique complète intégrant l’ensemble des questions scientifiques, en mettant en oeuvre des démarches de modélisation hydraulique et mécanique s’appuyant sur des données réelles disponibles sur un barrage en remblai. La démarche générale développée est appliquée à un barrage bien documenté servant de cas d’étude. Le modèle hydromécanique utilise la méthode des éléments finis et est développé avec le code élément finis Cast3M ouvert et compatible avec un usage en recherche scientifique. Le modèle développé permet le calcul du facteur de sécurité de l’ouvrage par la méthode de réduction de paramètres en intégrant un champ de pressions interstitielles calculé en régime transitoire. La variabilité spatiale des paramètres des matériaux du remblai est modélisée à partir d’une analyse géostatistique des mesures de contrôle de compactage sous forme de champs aléatoires qui sont intégrés au modèle éléments finis. Un couplage mécano-fiabiliste entre le code de calcul Cast3M et le logiciel de fiabilité OpenTURNS permet au final de propager les incertitudes et d’évaluer la fiabilité de l’ouvrage. / Hydraulic works – dams and dikes – are risky civil engineering structures. Dramatic consequences in terms of human and material losses may be induced by their failure. Embankment dams represent an important part of the whole dams in France and the majority of dams worldwide, without considering the important lengths of fluvial and coastal dikes. The structural safety of such structures is traditionally evaluated with limit-state deterministic or semi-probabilistic methods. Nevertheless, French regulations regarding hydraulic works has recently evolved and now impose for all large dams the realization of risk assessment studies based on probabilistic approach. In this purpose, the principal objective of this thesis work is to develop a probabilistic approach to evaluate earth dam reliability concerning the sliding mechanism, which is one of the designing limit-state of such structures. Three scientific issues have to be treated for developing such approach: · elaboration of an hydro-mechanical model for the dam deterministic evaluation towards sliding mechanism; · probabilistic modeling of hydraulic and mechanical soil properties spatial variability; · mechanical-reliability coupling with integration of the spatial variability representations in the hydro-mechanical model. Some research studies already exist on these issues. However these works concern generally only a part of the general issue: the lack of global work is to be deplored. Our work proposes a global methodologic approach taking into account the whole scientific issues and applying hydraulic and mechanical modeling approaches based on real data available in the earth dam. The developed approach is then applied on a dam case study. Hydro-mechanical model uses finite element method and is developed with the user-free code Cast3M which is compatible for a research use. This code allows the safety factor calculation through the strength reduction technique with integration of the pore pressures field estimated in transient condition. The spatial variability of embankment properties is represented with random fields based on a geostatistical analysis of construction controls data. These random fields are then integrated into the finite element model. A coupling between the physical finite element code Cast3M and the reliability software OpenTURNS finally allows assessing the uncertainties propagation and the reliability evaluation of the studied dam.

The transport of contaminants in groundwater system is strongly influenced by various aquifer heterogeneity factors such as spatial aquifer heterogeneity of hydraulic conductivity and reactive substances distribution. The contaminants transport can be simulated by using numerical reactive transport models, and their fate can be possibly even predicted. Furthermore, reactive transport modeling is an essential tool to get a profound understanding of hydrological-geochemical complex processes and to make plausible predictions of assessment. The goal of this work is to improve our understanding of the groundwater contaminants fate and transport processes in heterogeneous aquifer systems, with a focus on nitrate problems. A large body of knowledge of the fate and transport of nitrogen species has been achieved by previous works, however, most previous models typically neglect the interrelation of physical and chemical aquifer heterogeneities on the contaminant fate and redox transformation, which is required for predicting the movement and behavior of nitrate and quantifying the impact of uncertainty of numerical groundwater simulation, and which motivates this study. The main research questions which are answered in this work are how aquifer heterogeneity influences on the nitrate fate and transport and then, what is the most influential aquifer heterogeneity factor must be considered. Among the various type of aquifer heterogeneity, physical and chemical aquifer heterogeneities are considered. The first part of the work describes groundwater flow system and hydrochemical characteristics of the study area (Hessian Ried, Germany). Especially, data analyses are performed with the hydrochemical data to identify the major driving force for nitrate reduction in the study area. The second part of the work introduces a kinetic model describing nitrate removal by using numerical simulation. The resulting model reproduces nitrate reduction processes and captures the sequence of redox reactions. The third and fourth parts show the influence of physical and chemical aquifer heterogeneity with varying variance, correlation length scale, and anisotropy ratio. Heterogeneous aquifer systems are realized by using stochastic approach. Results, in short, show that the most influential aquifer heterogeneity factors could change over time. With abundant requisite electron donors, physical aquifer heterogeneity significantly influences the nitrate reduction while chemical aquifer heterogeneity plays a minor role. Increasing the spatial variability of the hydraulic conductivity increases the nitrate removal efficiency of the system in addition. If these conditions are reversed, nitrate removal efficiency varies by the spatial heterogeneity of the available initial electron donor. The results indicate that an appropriate characterization of the physical and chemical properties can be of significant importance to predict redox contamination transport and design long-term remediation strategies and risk assessment.

En France, les chutes constituent la première cause de mortalité chez les plus de 75 ans, et la seconde chez les plus de 65 ans. On estime qu'elle engendre un coût de 1 à 2 milliards d'euros par an pour la société. L'enjeu humain et socio-économique est colossal, sachant que le risque de chute est multiplié par 20 après une première chute, que le risque de décès est multiplié par 4 dans l'année qui suit une chute, que les chutes concernent 30% des personnes de plus de 65 ans et 50% des personnes de plus de 85 ans, et que l'on estime que d'ici 2050, plus de 30% de la population sera âgée de plus de 65 ans. Cette thèse propose un dispositif de détection de présence au sol se basant sur l'analyse de cartes de profondeurs acquises en temps réel, ainsi qu'une amélioration du dispositif proposé utilisant également un capteur thermique. Les cartes de profondeurs et les images thermiques nous permettent de nous affranchir des conditions d'illumination de la scène observée, et garantissent l'anonymat des personnes qui évoluent dans le champ de vision du dispositif. Cette thèse propose également différentes méthodes de détection du plan du sol dans une carte de profondeurs, le plan du sol constituant une référence géométrique nécessaire au dispositif proposé. Une enquête psychosociale a été réalisée, qui nous a permis d'évaluer l'acceptabilité a priori dudit dispositif. Cette enquête a démontré sa bonne acceptabilité, et a fourni des préconisations quant aux points d'amélioration et aux écueils à éviter. Enfin, une méthode de suivi d'objets dans une carte de profondeurs est proposée, un objectif à plus long terme consistant à mesurer l'activité des individus observés. / In France, fall is the first death cause for people aged 75 and more, and the second death cause for people aged 65 and more. It is considered that falls generate about 1 to 2 billion euros health costs per year. The human and social-economical issue is crucial, knowing that for the mentioned populations, fall risk is multiplied by 20 after a first fall; that the death risk is multiplied by 4 in the year following a fall; that per year, 30% of the people aged 65 and more and 50% of the people aged 85 and more are subject to falls; and that it is estimated that more than 30% of the French population whill be older than 65 years old by 2050. This thesis proposes a ground lying event detection device which bases on the real time analysis of depth maps, and also proposes an improvement of the device, which uses an additional thermal sensor. Depth maps and thermal images ensure the device is independent from textures and lighting conditions of the observed scenes, and guarantee that the device respects the privacy of those who pass into its field of view, since nobody can be recognized in such images. This thesis also proposes several methods to detect the ground plane in a depth map, the ground plane being a geometrical reference for the device. A psycho-social inquiry was conducted, and enabled the evaluation of the a priori acceptability of the proposed device. This inquiry demonstrated the good acceptability of the proposed device, and resulted in recommendations on points to be improved and on pitfalls to avoid. Last, a method to separate and track objects detected in a depth map is proposed, the measurement of the activity of observed individuals being a long term objective for the device.

Les nouvelles attentes vis-à-vis des nouveaux trains à grande vitesse sont nombreuses: on les voudrait plus rapides, plus confortables, plus stables, tout en étant moins consommateur d'énergie, moins agressif vis-à-vis des voies, moins bruyants… Afin d'optimiser la conception de ces trains du futur, il est alors nécessaire de pouvoir se baser sur une connaissance précise de l'ensemble des conditions de circulations qu'ils sont susceptibles de rencontrer au cours de leur cycle de vie. Afin de relever ces défis, la simulation a un très grand rôle à jouer. Pour que la simulation puisse être utilisée dans des perspectives de conception, de certification et d'optimisation de la maintenance, elle doit alors être tout à fait représentative de l'ensemble des comportements physiques mis en jeu. Le modèle du train, du contact entre les roues et le rail, doivent ainsi être validés avec attention, et les simulations doivent être lancées sur des ensembles d'excitations qui sont réalistes et représentatifs de ces défauts de géométrie. En ce qui concerne la dynamique, la géométrie de la voie, et plus particulièrement les défauts de géométrie, représentent une des principales sources d'excitation du train, qui est un système mécanique fortement non linéaire. A partir de mesures de la géométrie d'un réseau ferroviaire, un paramétrage complet de la géométrie de la voie et de sa variabilité semblent alors nécessaires, afin d'analyser au mieux le lien entre la réponse dynamique du train et les propriétés physiques et statistiques de la géométrie de la voie. Dans ce contexte, une approche pertinente pour modéliser cette géométrie de la voie, est de la considérer comme un champ aléatoire multivarié, dont les propriétés sont a priori inconnues. En raison des interactions spécifiques entre le train et la voie, il s'avère que ce champ aléatoire n'est ni Gaussien ni stationnaire. Ce travail de thèse s'est alors particulièrement concentré sur le développement de méthodes numériques permettant l'identification en inverse, à partir de mesures expérimentales, de champs aléatoires non Gaussiens et non stationnaires. Le comportement du train étant très non linéaire, ainsi que très sensible vis-à-vis de la géométrie de la voie, la caractérisation du champ aléatoire correspondant aux défauts de géométrie doit être extrêmement fine, tant du point de vue fréquentiel que statistique. La dimension des espaces statistiques considérés est alors très importante. De ce fait, une attention toute particulière a été portée dans ces travaux aux méthodes de réduction statistique, ainsi qu'aux méthodes pouvant être généralisées à la très grande dimension. Une fois la variabilité de la géométrie de la voie caractérisée à partir de données expérimentales, elle doit ensuite être propagée au sein du modèle numérique ferroviaire. A cette fin, les propriétés mécaniques d'un modèle numérique de train à grande vitesse ont été identifiées à partir de mesures expérimentales. La réponse dynamique stochastique de ce train, soumis à un très grand nombre de conditions de circulation réalistes et représentatives générées à partir du modèle stochastique de la voie ferrée, a été ainsi évaluée. Enfin, afin d'illustrer les possibilités apportées par un tel couplage entre la variabilité de la géométrie de la voie et la réponse dynamique du train, ce travail de thèse aborde trois applications / High speed trains are currently meant to run faster and to carry heavier loads, while being less energy consuming and still ensuring the safety and comfort certification criteria. In order to optimize the conception of such innovative trains, a precise knowledge of the realm of possibilities of track conditions that the train is likely to be confronted to during its life cycle is necessary. Simulation has therefore a big to play in this context. However, to face these challenges, it has to be very representative of the physical behavior of the system. From a general point of view, a railway simulation can be seen as the dynamic response of a non-linear mechanical system, the train, which is excited by a complex multivariate spatial function, the track geometry. Therefore, the models of the train, of the wheel/rail contact forces have thus to be fully validated and the simulations have to be raised on sets of excitations that are realistic and representative of the track geometry. Based on experimental measurements, a complete parametrization of the track geometry and of its variability would be of great concern to analyze the complex link between the train dynamics and the physical and statistical properties of the track geometry. A good approach to characterize this variability is to model the track geometry as a multivariate random field, for which statistical properties are only known through a set of independent realizations. Due to the specific interactions between the train and the track, this random field is neither stationary nor Gaussian. In order to propagate the track geometry variability to the train response, methods to identify in inverse, from a finite set of experimental data, the statistical properties of non-stationary and non-Gaussian random fields were analyzed in this thesis. The train behavior being very non-linear and very sensitive to the track geometry, the random field has to be described very precisely from frequency and statistical points of view. As a result, the statistical dimension of this random field is very high. Hence, a particular attention is paid in this thesis to statistical reduction methods and to statistical identification methods that can be numerically applied to the high dimensional case. Once the track geometry variability has been characterized from experimental data, it has to be propagated through the model. To this end, a normalized multibody model of a high speed train, whose mechanical parameters have been carefully identified from experimental measurements, has been made run on sets of realistic and representative running conditions. The commercial software Vampire was used to solve these dynamic equations. At last, three applications are proposed to illustrate to what extent such a railway stochastic modeling opens new possibilities in terms of virtual certification, predictive maintenance and optimization of the railway system

La reconnaissance des opinions d'un locuteur dans une interaction orale est une étape cruciale pour améliorer la communication entre un humain et un agent virtuel. Dans cette thèse, nous nous situons dans une problématique de traitement automatique de la parole (TAP) sur les phénomènes d'opinions dans des interactions orales spontanées naturelles. L'analyse d'opinion est une tâche peu souvent abordée en TAP qui se concentrait jusqu'à peu sur les émotions à l'aide du contenu vocal et non verbal. De plus, la plupart des systèmes récents existants n'utilisent pas le contexte interactionnel afin d'analyser les opinions du locuteur. Dans cette thèse, nous nous penchons sur ces sujet. Nous nous situons dans le cadre de la détection automatique en utilisant des modèles d’apprentissage statistiques. Après une étude sur la modélisation de la dynamique de l'opinion par un modèle à états latents à l’intérieur d'un monologue, nous étudions la manière d’intégrer le contexte interactionnel dialogique, et enfin d'intégrer l'audio au texte avec différents types de fusion. Nous avons travaillé sur une base de données de Vlogs au niveau d'un sentiment global, puis sur une base de données d'interactions dyadiques multimodales composée de conversations ouvertes, au niveau du tour de parole et de la paire de tours de parole. Pour finir, nous avons fait annoté une base de données en opinion car les base de données existantes n'étaient pas satisfaisantes vis-à-vis de la tâche abordée, et ne permettaient pas une comparaison claire avec d'autres systèmes à l'état de l'art.A l'aube du changement important porté par l’avènement des méthodes neuronales, nous étudions différents types de représentations: les anciennes représentations construites à la main, rigides mais précises, et les nouvelles représentations apprises de manière statistique, générales et sémantiques. Nous étudions différentes segmentations permettant de prendre en compte le caractère asynchrone de la multi-modalité. Dernièrement, nous utilisons un modèle d'apprentissage à états latents qui peut s'adapter à une base de données de taille restreinte, pour la tâche atypique qu'est l'analyse d'opinion, et nous montrons qu'il permet à la fois une adaptation des descripteurs du domaine écrit au domaine oral, et servir de couche d'attention via son pouvoir de clusterisation. La fusion multimodale complexe n'étant pas bien gérée par le classifieur utilisé, et l'audio étant moins impactant sur l'opinion que le texte, nous étudions différentes méthodes de sélection de paramètres pour résoudre ces problèmes. / 2588/5000Recognizing a speaker's opinions in an oral interaction is a crucial step in improving communication between a human and a virtual agent. In this thesis, we find ourselves in a problematic of automatic speech processing (APT) on opinion phenomena in natural spontaneous oral interactions. Opinion analysis is a task that is not often addressed in TAP that focused until recently on emotions using voice and non-verbal content. In addition, most existing legacy systems do not use the interactional context to analyze the speaker's opinions. In this thesis, we focus on these topics.We are in the context of automatic detection using statistical learning models. A study on modeling the dynamics of opinion by a model with latent states within a monologue, we study how to integrate the context interactional dialogical, and finally to integrate audio to text with different types of fusion. We worked on a basic Vlogs data at a global sense, and on the basis of multimodal data dyadic interactions composed of open conversations, at the turn of speech and word pair of towers. Finally, we annotated database in opinion because existing database were not satisfactory vis-à-vis the task addressed, and did not allow a clear comparison with other systems in the state art.At the dawn of significant change brought by the advent of neural methods, we study different types of representations: the ancient representations built by hand, rigid, but precise, and new representations learned statistically, and general semantics. We study different segmentations to take into account the asynchronous nature of multi-modality. Recently, we are using a latent state learning model that can adapt to a small database, for the atypical task of opinion analysis, and we show that it allows both an adaptation of the descriptors of the written domain to the oral domain, and serve as an attention layer via its clustering power. Complex multimodal fusion is not well managed by the classifier used, and audio being less impacting on opinion than text, we study different methods of parameter selection to solve these problems.