L'étude non destructive du bois à partir de scanners à rayons X nécessite d’imaginer de nouvelles solutions adaptées à l'analyse des images. Préoccupation à la fois de la recherche agronomique et du milieu industriel des scieries, la segmentation des nœuds de bois est un défi majeur en termes de robustesse aux spécificités de chaque espèce et aux conditions d'acquisition des images. Les travaux menés dans cette thèse permettent de proposer un processus de segmentation en deux phases. Il isole d'abord chaque nœud dans une zone réduite puis segmente le nœud unique de chaque zone. Les solutions proposées pour chaque phase permettent d'intégrer les connaissances sur l'organisation interne du tronc et les mécanismes inhérents à sa croissance, à travers des outils classiques du traitement et de l'analyse d'image. La première phase repose en grande partie sur un principe de détection du mouvement emprunté à l'analyse vidéo et revisité. Deux approches de segmentation sont ensuite proposées, considérant pour l'une les coupes tomographiques initiales, et pour l'autre de nouvelles coupes ré-échantillonnées pour chaque nœud, orthogonalement à sa trajectoire. L'intégralité du processus a été implémenté dans un logiciel dédié aussi bien à l'expérimentation et la validation de l'approche qu'aux échanges interdisciplinaires. Le support applicatif du bois souligne la capacité de spécialisation des algorithmes génériques du traitement et de l'analyse d'image, et la pertinence de l'intégration de connaissances a priori dans cette optique / The non-destructive study of wood from X-Ray CT scanners requires to imagine new solutions adapted to analysis of images. Relating both agronomic research and industrial sector of sawmills, segmentation of wood knots is a major challenge in terms of robustness to specificities of each species and to image acquisition conditions. The works carried out in this thesis allow to propose a segmentation process in two phases. It first isolates each knot in a reduced area then it segments the unique knot of each area. Proposed solutions for each phase allow to integrate knowledges about internal organization of trunk and mechanisms inherent to its growth, through classical tools of image analysis and processing. The first phase is essentially based on a movement detection principle borrowed from video analysis and revisited. Two segmentation approaches are then proposed, considering for one the initial CT slices and for the other news slices resampled for each knot orthogonally to its trajectory. The complete process has been implemented in a software dedicated both for experimentation and validation of approach, and to interdisciplinary dialogs. The applicative support of wood emphasizes the specialization abilities of generic image analysis and processing algorithms, and the relevance to integrate priori knowledges in this perspective

A economia mundial, cada vez mais competitiva, impulsiona o acelerado desenvolvimento tecnológico e conseqüentemente promove o crescente uso de Máquinas de Medir a Três Coordenadas (MM3Cs). A preferência por essas máquinas pode ser justificada pela sua grande flexibiÍidade e produtividade e é uma tentativa de melhorar a área de inspeção dimensional do processo produtivo. Um grande esforço no desenvolvimento de novas metodologias se realiza para se obter e modelar os erros de MM3Cs. Padrões para calibração de MM3Cs foram sugeridos e colocados em uso através dos diferentes anos, com a finalidade de utilizá-los em testes de aceitação e verificação periódica da incerteza de medição de MM3Cs. Novos artefatos para a calibração indireta de MM3Cs visam melhorar os procedimentos de calibração para uso em sistemas de compensação de erros. Diante do exposto acima, este trabalho tem como objetivo desenvolver um Procedimento de Calibração Indireta de MM3Cs com o Esquadro Mecânico de Esferas aliado a um Modelo Reduzido de Sintetização de Erros (MRSE) para uso em um Sistema de Compensação de Erros. O procedimento possibilita maior rapidez na obtenção dos valores e comportamentos dos erros quando comparado com outros procedimentos de calibração indireta. Aliado a um MRSE, o procedimento proposto tem como vantagem o uso de um único artefato para medir todos os termos das componentes do erro volumétrico, nas direções X, Y e Z de uma MM3C. / The increasingly competitive world economy promotes the technological development and, as a consequence, the growing use of Coordinate Measuring Machines (CMMs). The predilection for CMMs can be justified by their great flexibility and productivity and is also an attempt to improve the dimerisional inspection quality of the productive process. Considerable effort was made on the development of new methodologies to obtain and model CMMs errors. Calibration standards were suggested and employed for years aiming at acceptance tests and periodic verification of measurement uncertainty of CMMs. The development of new artefacts for CMM indirect calibration seeks to enhance calibration procedures for error compensation systems. Therefore, this work aims at the development of an indirect CMM calibration procedure using a mechanical ball square and the Error Synthesization Reduced Model (ESRM) for an error compensation system. The procedure allows faster obtaining of error values and behaviour if compared with other indirect calibration procedures. The proposed procedure, along with the ESRM, is advantageous since only one artefact is employed to measure all the terms of the volumetric error components m directions X, Y and Z of the CMM.

Le flot de Ricci est une équation aux dérivées partielles qui régit l’évolution d’une métrique riemannienne dépendant d’un paramètre de temps sur une variété différentielle. D’abord introduit et étudié par R. Hamilton, il est à l’origine de la solution de la conjecture de géométrisation des variétés compactes de dimension 3 par G. Perelman en 2001. La théorie classique concernant l’existence en temps court des solutions, due à Hamilton et à Shi, garantit (en dimension quelconque) l’existence d’un flot soit sur une variété compacte, soit lorsque la métrique initiale est complète avec une borne sur la norme du tenseur de courbure. En l’absence de cette borne, on conjecture qu’on peut trouver, à partir de la dimension 3, des données initiales pour lesquelles il n’existe pas de solution. Dans cette thèse, on démontre des théorèmes d’existence en temps court du flot sous des hypothèses plus faibles qu’une borne sur la norme du tenseur de courbure. Pour cela, on introduit une construction générale qui, pour une métrique riemannienne g quelconque sur une variété M, pas nécessairement complète, permet de produire une solution de l’équation du flot sur un domaine ouvert D de l’espace-temps M * [0,T] qui contient la tranche de temps initiale, avec g pour donnée initiale. On montre ensuite que sous des hypothèses adaptées sur la métrique g, on contrôle la forme du domaine D. En particulier, lorsque la métrique g est complète, D contient un ensemble de la forme M * [0,t], avec t>0, ce qui revient à dire qu’il existe un flot au sens classique dont la donnée initiale est g. Les « hypothèses adaptées » qui conduisent à des théorèmes d’existence sont de trois types. Dans tout les cas, on suppose une minoration uniforme du volume des boules de rayon au plus 1, à quoi on ajoute : a) en dimension 3, une minoration du tenseur de Ricci, b) en dimension n, une minoration d’une notion de courbure dite « courbure isotrope I » ou bien c) en dimension n, une borne sur la norme du tenseur de Ricci et une hypothèse qui garantit la proximité au sens métrique des boules de rayon au plus 1 avec une boule de même rayon dans un espace métrique obtenu comme le produit cartésien d’un espace de dimension 3 et d’un facteur euclidien de dimension n-3. De plus, avec ces résultats d’existence viennent des estimations sur les propriétés de régularisation du flot quantifiées en fonction des hypothèses sur la donnée initiale. La possibilité ainsi offerte de régulariser, globalement ou localement, pour un temps et avec des estimations quantifiés, une métrique initiale a des conséquence sur les espaces métriques singuliers obtenus comme limites, pour la distance de Gromov-Hausdorff, de suites de variétés satisfaisant uniformément aux conditions a), b) ou c). En effet, des théorèmes de compacité classiques pour le flot de Ricci permettent d’extraire un flot limite, étant donnée une suite de métriques initiales satisfaisant uniformément à ces hypothèses, et possédant donc toutes un flot pour un temps contrôlé. Lorsque les métriques en question approchent, pour la topologie de Gromov-Hausdorff, un espace singulier, cette solution limite s’interprète comme un flot régularisant l’espace singulier en question, et son existence contraint la topologie de cet espace singulier. / The Ricci Flow is a partial differential equation governing the evolution of a Riemannian metric depending on a time parameter t on a differential manifold. It was first introduced and studied by R. Hamilton, and eventually led to the solution of the Geometrization conjecture for closed three-dimensional manifolds by G. Perelman in 2001. The classical short-time existence theory for the Ricci Flow, due to Hamilton and Shi, asserts, in any dimension, the existence of a flow starting from any initial metric when the underlying manifold in compact, or for any complete initial metric with a bound on the norm of the curvature tensor otherwise. In the absence of such a bound, though, the conjecture is that starting from dimension 3 one can find such initial data for which there is no solution. In this thesis, we prove short-time existence theorems under hypotheses weaker than a bound on the norm of the curvature tensor. To do this, we introduce a general construction which, for any Riemannian metric g (not necessarily complete) on a manifold M, allows us to produce a solution to the equation of the flow on an open domain D of the space-time M * [0,T] which contains the initial time slice, with g as an initial datum. We proceed to show that under suitable hypotheses on g, one can control the shape of the domain D, so that in particular, D contains a subset of the form M * [0,t] with t>0 if g is complete. By « suitable hypothesis », we mean one of the following. In any case, we assume a lower bound on the volume of balls of radius at most 1, plus a) in dimension 3, a lower bound on the Ricci tensor, b) in dimension n, a lower bound on the so-called « isotropic curvature I » or c) in dimension n, a bound on the norm of the Ricci tensor, as well as a hypothesis which garanties the metric proximity of every ball of radius at most $1$ with a ball of the same radius in a metric product between a three-dimensional metric space and a $n-3$ dimensional Euclidian factor. Moreover, with these existence results come estimates on the existence time and regularization properties of the flow, quantified in term of the hypotheses on the initial data. The possibility to regularize metrics, locally or globally, with such estimates has consequences in terms of the metric spaces obtained as limits, in the Gromov-Hausdorff topology, of sequences of manifolds uniformly satisfying a), b) or c). Indeed, the classical compactness theorems for the Ricci Flow allow for the extraction of a limit flow for any sequence of initial metrics uniformly satisfying the hypotheses and thus possessing a flow for a controlled amount of time. In the case when these metrics approach a singular space in the Gromov-Hausdorff topology, such a limit solution can be interpreted as a flow regularizing the singular limit space, the existence of which puts constraints on the topology of this space.

This dissertation considers several problems related to Ricci flow, including the existence and behavior of solutions. The first goal is to obtain explicit, coordinate-based descriptions of Ricci flow solutions--especially those corresponding to Ricci solitons--on two classes of nilpotent Lie groups. On the odd-dimensional classical Heisenberg groups, we determine the asymptotics of Ricci flow starting at any metric, and use Lott's blowdown method to demonstrate convergence to soliton metrics. On the groups of real unitriangular matrices, which are more complicated, we describe the solitons and corresponding solutions using a suitable ansatz. Next, we consider solsolitons involving the nilsolitons in the Heisenberg case above. This uses work of Lauret, which characterizes solsolitons as certain extensions of nilsolitons, and work of Will, which demonstrates that the space of solsolitons extensions of a given nilsoliton is parametrized by the quotient of a Grassmannian by a finite group. We determine these spaces of solsoliton extensions of Heisenberg nilsolitons, and we also explicitly describe many-parameter families of these solsolitons in dimensions greater than three. Finally, we explore Ricci flow coupled with harmonic map flow, both as it arises naturally in certain bundle constructions related to Ricci flow and as a geometric flow in its own right. In the first case, we generalize a theorem of Knopf that demonstrates convergence and stability of certain locally R[superscript N]-invariant Ricci flow solutions. In the second case, we prove a version of Hamilton's compactness theorem for the coupled flow, and then generalize it to the category of etale Riemannian groupoids. We also provide a detailed example of solutions to the flow on the three-dimensional Heisenberg group. / text

This thesis examines the elegant theory of polar and Legendre duality, and its potential use in convex geometry and geometric analysis. It derives a theorem of polar - Legendre duality for all convex bodies, which is captured in a commutative diagram. A geometric flow on a convex body induces a distortion on its polar dual. In general these distortions are not flows defined by local curvature, but in two dimensions they do have similarities to the inverse flows on the original convex bodies. These ideas can be extended to higher dimensions. Polar - Legendre duality can also be used to examine Mahler's Conjecture in convex geometry. The theory presents new insight on the resolved two-dimensional problem, and presents some ideas on new approaches to the still open three dimensional problem.

A economia mundial, cada vez mais competitiva, impulsiona o acelerado desenvolvimento tecnológico e conseqüentemente promove o crescente uso de Máquinas de Medir a Três Coordenadas (MM3Cs). A preferência por essas máquinas pode ser justificada pela sua grande flexibiÍidade e produtividade e é uma tentativa de melhorar a área de inspeção dimensional do processo produtivo. Um grande esforço no desenvolvimento de novas metodologias se realiza para se obter e modelar os erros de MM3Cs. Padrões para calibração de MM3Cs foram sugeridos e colocados em uso através dos diferentes anos, com a finalidade de utilizá-los em testes de aceitação e verificação periódica da incerteza de medição de MM3Cs. Novos artefatos para a calibração indireta de MM3Cs visam melhorar os procedimentos de calibração para uso em sistemas de compensação de erros. Diante do exposto acima, este trabalho tem como objetivo desenvolver um Procedimento de Calibração Indireta de MM3Cs com o Esquadro Mecânico de Esferas aliado a um Modelo Reduzido de Sintetização de Erros (MRSE) para uso em um Sistema de Compensação de Erros. O procedimento possibilita maior rapidez na obtenção dos valores e comportamentos dos erros quando comparado com outros procedimentos de calibração indireta. Aliado a um MRSE, o procedimento proposto tem como vantagem o uso de um único artefato para medir todos os termos das componentes do erro volumétrico, nas direções X, Y e Z de uma MM3C. / The increasingly competitive world economy promotes the technological development and, as a consequence, the growing use of Coordinate Measuring Machines (CMMs). The predilection for CMMs can be justified by their great flexibility and productivity and is also an attempt to improve the dimerisional inspection quality of the productive process. Considerable effort was made on the development of new methodologies to obtain and model CMMs errors. Calibration standards were suggested and employed for years aiming at acceptance tests and periodic verification of measurement uncertainty of CMMs. The development of new artefacts for CMM indirect calibration seeks to enhance calibration procedures for error compensation systems. Therefore, this work aims at the development of an indirect CMM calibration procedure using a mechanical ball square and the Error Synthesization Reduced Model (ESRM) for an error compensation system. The procedure allows faster obtaining of error values and behaviour if compared with other indirect calibration procedures. The proposed procedure, along with the ESRM, is advantageous since only one artefact is employed to measure all the terms of the volumetric error components m directions X, Y and Z of the CMM.

In this thesis, the stability of the family of subextremal Kerr-Newman space- times is studied in the case of linear scalar perturbations. That is, nondegenerate energy bounds (NEB) and integrated local energy decay (ILED) results are proved for solutions of the wave equation on the domain of outer communications. The main obstacles to the proof of these results are superradiance, trapping and their interaction. These difficulties are surmounted by localising solutions of the wave equation in phase space and applying the vector field method. Miraculously, as in the Kerr case, superradiance and trapping occur in disjoint regions of phase space and can be dealt with individually. Trapping is a high frequency obstruction to the proof whereas superradiance occurs at both high and low frequencies. The construction of energy currents for superradiant frequencies gives rise to an unfavourable boundary term. In the high frequency regime, this boundary term is controlled by exploiting the presence of a large parameter. For low superradiant frequencies, no such parameter is available. This difficulty is overcome by proving quantitative versions of mode stability type results. The mode stability result on the real axis is then applied to prove integrated local energy decay for solutions of the wave equation restricted to a bounded frequency regime. The (ILED) statement is necessarily degenerate due to the trapping effect. This implies that a nondegenerate (ILED) statement must lose differentiability. If one uses an (ILED) result that loses differentiability to prove (NEB), this loss is passed onto the (NEB) statement as well. Here, the geometry of the subextremal Kerr-Newman background is exploited to obtain the (NEB) statement directly from the degenerate (ILED) with no loss of differentiability.

Cette thèse porte sur l'analyse visuelle de scènes intérieures capturées par des caméras de profondeur dans le but de convertir leurs données en information de haut niveau sur la scène. Elle explore l'application d'outils d'analyse géométrique 3D à des données visuelles de profondeur en termes d'amélioration de qualité, de recalage et de consolidation. En particulier, elle vise à montrer comment l'abstraction de formes permet de générer des représentations légères pour une analyse rapide avec des besoins matériels faibles. Cette propriété est liée à notre objectif de concevoir des algorithmes adaptés à un fonctionnement embarqué en temps réel dans le cadre d'appareils portables, téléphones ou robots mobiles. Le contexte de cette thèse est l'exécution d'un procédé d’interaction 3D temps réel sur un appareil mobile. Cette exécution soulève plusieurs problématiques, dont le placement de zones d'interaction 3D par rapport à des objets environnants réels, le suivi de ces zones dans l'espace lorsque le capteur est déplacé ainsi qu'une utilisation claire et compréhensible du système par des utilisateurs non experts. Nous apportons des contributions vers la résolution de ces problèmes pour montrer comment l'abstraction géométrique de la scène permet une localisation rapide et robuste du capteur et une représentation efficace des données fournies ainsi que l'amélioration de leur qualité et leur consolidation. Bien que les formes géométriques simples ne contiennent pas autant d'information que les nuages de points denses ou les ensembles volumiques pour représenter les scènes observées, nous montrons qu’elles constituent une approximation acceptable et que leur légèreté leur donne un bon équilibre entre précision et performance. / This PhD thesis focuses on the problem of visual scene analysis captured by commodity depth sensors to convert their data into high level understanding of the scene. It explores the use of 3D geometry analysis tools on visual depth data in terms of enhancement, registration and consolidation. In particular, we aim to show how shape abstraction can generate lightweight representations of the data for fast analysis with low hardware requirements. This last property is important as one of our goals is to design algorithms suitable for live embedded operation in e.g., wearable devices, smartphones or mobile robots. The context of this thesis is the live operation of 3D interaction on a mobile device, which raises numerous issues including placing 3D interaction zones with relation to real surrounding objects, tracking the interaction zones in space when the sensor moves and providing a meaningful and understandable experience to non-expert users. Towards solving these problems, we make contributions where scene abstraction leads to fast and robust sensor localization as well as efficient frame data representation, enhancement and consolidation. While simple geometric surface shapes are not as faithful as heavy point sets or volumes to represent observed scenes, we show that they are an acceptable approximation and their light weight makes them well balanced between accuracy and performance.

As pontes estaiadas são exemplos de estruturas esbeltas e flexíveis onde a capacidade de utilização dos materiais tem grande importância. Neste sentido, para garantir a melhor utilização dos materiais envolvidos (aço e concreto, por exemplo), é preciso determinar as forças de protensão aplicadas aos cabos. A melhor distribuição dos momentos fletores no tabuleiro de ponte é aquela obtida com uma viga contínua. Pontes estaiadas fornecem apoios elásticos ao tabuleiro. O presente trabalho emprega o método da anulação dos deslocamentos, MAD, para obter as forças axiais a que os cabos estarão submetidos de modo a aproximar o comportamento do tabuleiro ao de uma viga contínua. O método MAD. proporciona uma estrutura economicamente mais viável. O código computacional desenvolvido realiza análises estática e modal por meio do método dos elementos finitos, MEF. A análise estática utilizada é a não linear geométrica, considerando as não linearidades do efeito de catenária do cabo, e dos elementos submetidos à compressão. O material é assumido no campo do regime elástico linear. A ponte é modelada por elementos de treliça plana com módulo de elasticidade de Dischinger, para simular os cabos, e elementos de pórtico plano para os elementos do tabuleiro e da torre. O carregamento da estrutura considera a atuação apenas do peso-próprio dos elementos estruturais. O código computacional desenvolvido permite, também, a análise modal da estrutura a fim de determinar suas frequências naturais e modos de vibração. A análise modal pode ser realizada com a matriz de massa concentrada, ou consistente. Em relação à matriz de rigidez, a análise modal da estrutura pode utilizar a matriz de rigidez linear, para uma análise de vibrações livres, ou a matriz de rigidez tangente para as análises de vibração sob tensões iniciais. Exemplos encontrados na literatura são resolvidos com o código computacional desenvolvido para verificação e validação. / The cable-stayed bridges are examples of slender and flexible where the usability of the materials is very important structures. In this sense, to ensure the best use of the materials involved (steel and concrete, for example), one must determine the forces applied to the prestressing cables. A better distribution of the bending moments in the bridge deck is obtained with a continuous beam. Cable-stayed bridges provide elastic support to the deck. This work employs the zero displacement method, ZDM, to determine the axial forces that the cables will be subjected to in order to approximate the behavior of the deck to the one as a continuous beam. The ZDM method provides an economically viable structure. The computational code performs static and modal analysis, which are performed by using the finite element method, FEM. The static analysis is a nonlinear geometric analysis which considers the nonlinearities of the cable sag, and the compression effects on the elements. The material is assumed in the field of linear elastic regime. The bridge is modeled by elements of plane truss with Dischingers elasticity module, to simulate cables and plane frame elements for the deck and the tower elements. The structure is subjected to self-weight of the elements. The computer code developed also performs the modal analysis of the structure to determine their natural frequencies and mode shapes. The modal analysis can be carried out with the concentrated or consistent mass matrix. In relation to the stiffness matrix, modal analysis of the structure may use a linear stiffness matrix for analysis of free vibration analysis or the tangent stiffness matrix for the analysis of vibration under initial stress. Examples in the literature are solved with the computational code developed for verification and validation.

As pontes estaiadas são exemplos de estruturas esbeltas e flexíveis onde a capacidade de utilização dos materiais tem grande importância. Neste sentido, para garantir a melhor utilização dos materiais envolvidos (aço e concreto, por exemplo), é preciso determinar as forças de protensão aplicadas aos cabos. A melhor distribuição dos momentos fletores no tabuleiro de ponte é aquela obtida com uma viga contínua. Pontes estaiadas fornecem apoios elásticos ao tabuleiro. O presente trabalho emprega o método da anulação dos deslocamentos, MAD, para obter as forças axiais a que os cabos estarão submetidos de modo a aproximar o comportamento do tabuleiro ao de uma viga contínua. O método MAD. proporciona uma estrutura economicamente mais viável. O código computacional desenvolvido realiza análises estática e modal por meio do método dos elementos finitos, MEF. A análise estática utilizada é a não linear geométrica, considerando as não linearidades do efeito de catenária do cabo, e dos elementos submetidos à compressão. O material é assumido no campo do regime elástico linear. A ponte é modelada por elementos de treliça plana com módulo de elasticidade de Dischinger, para simular os cabos, e elementos de pórtico plano para os elementos do tabuleiro e da torre. O carregamento da estrutura considera a atuação apenas do peso-próprio dos elementos estruturais. O código computacional desenvolvido permite, também, a análise modal da estrutura a fim de determinar suas frequências naturais e modos de vibração. A análise modal pode ser realizada com a matriz de massa concentrada, ou consistente. Em relação à matriz de rigidez, a análise modal da estrutura pode utilizar a matriz de rigidez linear, para uma análise de vibrações livres, ou a matriz de rigidez tangente para as análises de vibração sob tensões iniciais. Exemplos encontrados na literatura são resolvidos com o código computacional desenvolvido para verificação e validação. / The cable-stayed bridges are examples of slender and flexible where the usability of the materials is very important structures. In this sense, to ensure the best use of the materials involved (steel and concrete, for example), one must determine the forces applied to the prestressing cables. A better distribution of the bending moments in the bridge deck is obtained with a continuous beam. Cable-stayed bridges provide elastic support to the deck. This work employs the zero displacement method, ZDM, to determine the axial forces that the cables will be subjected to in order to approximate the behavior of the deck to the one as a continuous beam. The ZDM method provides an economically viable structure. The computational code performs static and modal analysis, which are performed by using the finite element method, FEM. The static analysis is a nonlinear geometric analysis which considers the nonlinearities of the cable sag, and the compression effects on the elements. The material is assumed in the field of linear elastic regime. The bridge is modeled by elements of plane truss with Dischingers elasticity module, to simulate cables and plane frame elements for the deck and the tower elements. The structure is subjected to self-weight of the elements. The computer code developed also performs the modal analysis of the structure to determine their natural frequencies and mode shapes. The modal analysis can be carried out with the concentrated or consistent mass matrix. In relation to the stiffness matrix, modal analysis of the structure may use a linear stiffness matrix for analysis of free vibration analysis or the tangent stiffness matrix for the analysis of vibration under initial stress. Examples in the literature are solved with the computational code developed for verification and validation.