The purpose of this study is to develop an automatic facial feature extraction system, which is able to identify the detailed shape of eyes, eyebrows and mouth from facial images. The developed system not only extracts the location information of the features, but also estimates the parameters pertaining the contours and parts of the features using parametric deformable templates approach. In order to extract facial features, deformable models for each of eye, eyebrow, and mouth are developed. The development steps of the geometry, imaging model and matching algorithms, and energy functions for each of these templates are presented in detail, along with the important implementation issues. In addition, an eigenfaces based multi-scale face detection algorithm which incorporates standard facial proportions is implemented, so that when a face is detected the rough search regions for the facial features are readily available. The developed system is tested on JAFFE (Japanese Females Facial Expression Database), Yale Faces, and ORL (Olivetti Research Laboratory) face image databases. The performance of each deformable templates, and the face detection algorithm are discussed separately.

This thesis explores two possible ways of detecting soft plaque present in the coronary arteries, using CTA imagery. The coronary arteries are vessels that supply oxidized blood to the cardiac muscle and are thus important for the proper functioning of heart. Cholesterol or reactive oxygen species from cigarette smoke and other toxins may get adhered to the walls of coronary arteries and trigger chronic inflammation that leads to formation of the soft plaque. When the soft plaque grows bigger in volume, it occludes the blood flow to the cardiac muscle and finally results in ischemic heart attack. Moreover, smaller plaque can easily rupture due to the blood flow in arteries and can result in complications such as stroke. Hence there is a need to detect the soft plaque using non-invasive or minimally invasive techniques. In CTA imagery, the cardiac muscle appears as a dark gray color, while the blood appears as dull white color and the the calcified plaque appears as bright white. The soft plaque has an intensity which falls between the intensity level of the blood and cardiac muscle, making it difficult to directly segment the soft plaque using standard segmentation methods. Soft plaque in its advanced stages forms a concavity in the blood lumen. A watershed based segmentation method was used to detect the presence of this concavity which in turn identifies the location of the soft plaque. For segmenting the soft plaque at its earlier stages, a novel segmentation technique was used. In this technique the surface is evolved based on a region-based energy calculated in the local neighborhood around each point on the evolving surface. This method seems to be superior to the watershed based segmentation method in detecting smaller plaque deposits.

Devido à variação na qualidade e ao ruído nas imagens médicas, a aplicação de técnicas tradicionais de segmentação é geralmente ineficiente. Nesse sentido, apresenta-se um novo algoritmo a partir de duas técnicas: o modelo de contornos deformáveis por fluxo do vetor gradiente (GVF deformable contours) e a técnica de espaço de escalas utilizando o processo de difusão. Assim, foi realizada uma revisão bibliográfica dos modelos que trabalham com os contornos deformáveis, os quais foram classificados em modelos paramétricos e geométricos. Entre os modelos paramétricos foi escolhido o modelo de contornos deformáveis por fluxo do vetor gradiente (GVF). Esta aproximação oferece precisão na representação de estruturas biológicas não observada em outros modelos. Desta forma, o algoritmo apresentado mapeia as bordas (edge map) e aperfeiçoa a condução da deformação utilizando uma técnica baseada em operações recursivas. Com este cálculo apoiado no comportamento de espaço de escalas, obtem-se a localização e correção de sub-regiões do edge map que perturbam a deformação. Por outro lado, é incorporada uma nova característica que permite ao algoritmo realizar atividades de classificação. O algoritmo consegue determinar a presença ou ausência de um objeto de interesse utilizando um valor mínimo de deformação. O algoritmo é validado através do tratamento de imagens sintéticas e médicas comparando os resultados com os obtidos no modelo tradicional de contornos deformáveis GVF. / Due to the variation of the quality and noise in medical images, the classic image segmentation techniques are usually ineffective. In this work, we present a new algorithm that is composed of two techniques: the gradient vector flow deformable contours (GVF) and the scale-space technique using a diffusion process. A bibliographical revision of the models that work with deformable contours was accomplished, they were classified in parametric and geometric models. Among the parametric models the gradient vector flow deformable contours (GVF) was chosen. This approach offers precision in the representation of biological structures where other models does not. Thus, the algorithm improves the edge map to guide the deformation using recursive operations. With this estimation based on the behavior of the scale-space techniques it is realized, the localization and correction of sub-areas of the edge map that disturb the deformation. On the other hand, it was incorporated a new characteristic that allows the algorithm to accomplish classification activities. That is, the algorithm determines the presence or absence of a target object using a minimal deformation area. Our method was validated on both, simulated images and medical images making a comparison with the traditional GVF deformable contours.

Just how far is it possible to make learning of new parts for recognition and robot picking autonomous? This thesis initially gives the prerequisites for the steps in learning and calibration that are to be automated. Among these tasks are to select a suitable part model from numerous candidates with the help of a new part segmenter, as well as computing the spatial extent of this part, facilitating robotic collision handling. Other tasks are to analyze the part model in order to highlight correct and suitable edge segments for increasing pattern matching certainty, and to choose appropriate acceptance levels for pattern matching. Furthermore, tasks deal with simplifying camera calibration by analyzing the calibration pattern, as well as compensating for differences in perspective at great depth variations, by calculating the centre of perspective of the image. The image processing algorithms created in order to solve the tasks are described and evaluated thoroughly. This thesis shows that simplification of steps of learning and calibration, by the help of advanced image processing, really is possible.

La qualité d’une prise de vue est un point incontournable dans la résolution des problèmes d’imagerie. Un capteur non adapté, un éclairage non contrôlé, ou des conditions variables de la scène observée peuvent être à l’origine de problèmes très difficiles à surmonter. Nous présentons différentes méthodes de traitement d’image permettant de prendre en compte au mieux ces conditions de prise de vue instables. Les approches que nous proposons sont définies dans le cadre du modèle LIP (Logarithmic Image Processing). Dans une première partie, nous nous intéressons à des notions de contraste : le contraste LIP additif et un nouveau contraste LIP multiplicatif, ainsi qu’à leurs métriques associées. De nouveaux outils de traitement basés sur ces notions sont ensuite définis : seuillage, détecteur de contours, reconnaissance de modèle. L’utilisation de ces notions de contraste confère à ces algorithmes la capacité des contrastes LIP à s’adapter à différents types d’images mal conditionnées. Nous proposons ensuite de nouvelles techniques de correction de dynamique d’images en exploitant les opérations LIP. Différentes corrections globales et locales sont présentées ainsi que leurs applications directes : correction de dérive d’éclairement pour du contrôle industriel ou amélioration d’image pour de la visualisation. Nous obtenons notamment une méthode de correction locale dont les résultats se rapprochent de ceux de certaines techniques de tone mapping. En comparaison, notre technique s’avère simple, rapide (temps réel à 30 images par seconde) et réaliste car basée sur une interprétation physique de la problématique / The quality of image acquisitions is crucial in the resolution of imaging problems. Troubles during acquisiton can lead to unstability for image processing algorithms. We propose different methods (thresholding techniques, contour detection, pattern matching) based on new metrics and contrasts in the LIP context. The LIP (Logarithmic Image Processing) model is recognized as an efficient framework to process images acquired in transmitted light and to take into account the human visual system. LIP operations are also useful to simulate varitations of image parameters in situation of reflected light. Finally, we propose new methods of global and local dynamic enhancement in the LIP framework like a real time and realistic local dynamic correction that brings results close to those obtained by certain tone mapping methods

Cette thèse s’inscrit dans le cadre de l’analyse des besoins de personnes malvoyantes dans le but de développer de nouvelles solutions d’assistance visuelle en utilisant le principe de réalité augmentée. Le principal objectif de ces travaux était le développement d’un dispositif d’assistance visuelle pour faciliter la mobilité de personnes malvoyantes grâce à un algorithme qui en temps réel adapte sa réponse en fonction des besoins de l’utilisateur et de la situation dans laquelle il évolue. Le premier postulat de cette thèse est d’utiliser l’acuité résiduelle fonctionnelle des malvoyants afin de leur permettre de percevoir un certain nombre d’informations. Le second postulat est d’utiliser les seules informations pertinentes, utiles à un malvoyant, afin d’améliorer sa perception du monde dans lequel il évolue. Le premier axe de recherche a consisté à analyser les besoins des malvoyants afin de recenser, les indices visuels utiles à rehausser afin d’améliorer la perception des malvoyants. Cette étude a été réalisée en immersion et une enquête précise a été menée auprès de malvoyants. Cette étude permet de mieux comprendre les besoins des malvoyants, les méthodes qu’ils utilisent ainsi que leurs usages des dispositifs actuels. Le deuxième axe de recherche a consisté à concevoir et à mettre en oeuvre un algorithme de traitement d’images répondant aux besoins identifiés. Un outil interactif a été développé qui permet de sélectionner la quantité de contours en fonction de la situation. Ce détecteur de contour optimise sa détection afin d’obtenir des contours plus exploitables pour les malvoyants, comparés aux résultats des autres méthodes de l’état de l’art. Cet outil a été testé et validé auprès d’une population donnée de personnes malvoyantes. Ces tests réalisés sur écran ont démontré que les contours constituent des indices visuels très importants qui peuvent être d’une grande aide dans l’analyse d’une scène. Le troisième axe de travail a consisté au développement d’un système de réalité augmentée permettant d’utiliser ce détecteur de contours dans un contexte de mobilité. L’analyse des besoins des malvoyants vis-à-vis d’un tel dispositif nous a permis de définir les caractéristiques indispensables pour le système de calcul, la caméra et le dispositif d’affichage. Concernant le système d’affichage, l’analyse des besoins des malvoyants en termes de profil, de couleur, de largeur des contours, nous a permis de définir les paramètres nécessaires pour un affichage qui répond à ces besoins. Il a également été pris en compte dans les développements les paramètres physiologiques de chaque observateur afin que les images affichées soient visuellement confortables / This thesis was done in the context of visual impaired needs analysis with the goal of new augmented reality visual assistance device development. The main goal of this work was the development of assistive device for mobility of visual impaired people by the use real-time image processing algorithm which take into account user needs and the context. The first postulate is the use of residual visual acuity of visual impaired people to allow them to see more information. The second postulate is the use only most useful information for visual impaired people for the understanding of the situation where are they. The first research axis was an analysis of needs of visual impaired to identify visual cues, which one we need to enhance to help them. To do this, I have done an immersive stage into a re-adaptation clinic for visually impaired. I have also done a questionnaire to identify and understand how current products answers partially to their needs. The second research axis was the development of a new image processing to answer identified needs. I have developed an interactive tools that allow the user to select the amount of edges for each situation, context. This detector optimizes its result to provide better results than others state of the art evaluated methods. This tool was evaluated and validated during visual test by visual impaired people. We have demonstrated that edges are one of the most important visual cues for visual impaired people to analyze a scene. The third research axis was the development of the augmented reality system to use my tool in a context of mobility. To do this, I have analyzed the needs of visual impaired people for this type of devices. I have also defined main requirements of the desired device for the processing unit, the camera and the display module. For display module, I have analyzed physiologic needs of visually impaired, in terms of profile, colors, and thickness to answer their needs. I have also developed an image generation model which takes into account many physiologic parameters of each observer to create visually comfortable images

The PhD thesis deals with modern methods of image processing, especially image segmentation, classification and evaluation of parameters. It is focused primarily on processing medical images of soft tissues obtained by magnetic resonance tomography (MR) and microscopic images of tissues. It is easy to describe edges of the sought objects using of segmented images. The edges found can be useful for further processing of monitored object such as calculating the perimeter, surface and volume evaluation or even three-dimensional shape reconstruction. The proposed solutions can be used for the classification of healthy/unhealthy tissues in MR or other imaging. Application examples of the proposed segmentation methods are shown in this thesis. Research in the area of image segmentation is focused on methods based on solving partial differential equations. This is a modern method for image processing, often called the active contour method. It is of great advantage in the segmentation of real images degraded by noise with fuzzy edges and transitions between objects. The results of the thesis are methods proposed for automatic image segmentation and classification.

The dissertation thesis deals with 3D reconstruction of the temporomandibular joint from 2D slices of tissue obtained by magnetic resonance. The current practice uses 2D MRI slices in diagnosing. 3D models have many advantages for the diagnosis, which are based on the knowledge of spatial information. Contemporary medicine uses 3D models of tissues, but with the temporomandibular joint tissues there is a problem with segmenting the articular disc. This small tissue, which has a low contrast and very similar statistical characteristics to its neighborhood, is very complicated to segment. For the segmentation of the articular disk new methods were developed based on the knowledge of the anatomy of the joint area of the disk and on the genetic-algorithm-based statistics. A set of 2D slices has different resolutions in the x-, y- and z-axes. An up-sampling algorithm, which seeks to preserve the shape properties of the tissue was developed to unify the resolutions in the axes. In the last phase of creating 3D models standard methods were used, but these methods for smoothing and decimating have different settings (number of polygons in the model, the number of iterations of the algorithm). As the aim of this thesis is to obtain the most precise model possible of the real tissue, it was necessary to establish an objective method by which it would be possible to set the algorithms so as to achieve the best compromise between the distortion and the model credibility achieve.

Cette thèse se présente sous la forme d’une enquête menée dans une société littorale de Normandie, au XVIIIe siècle, afin d’appréhender les liens qui unissent les gens de mer aux autres habitants du rivage, mais aussi de déterminer ce qui peut les séparer, dans un monde a priori tourné vers la mer. L’étude a pour cadre la rade de la Hougue, sur la côte est du Cotentin, qui abrite la population de Saint-Vaast-la-Hougue, objet de nos investigations.Un chapitre liminaire présente les conditions d’ouverture vers le large de ce bourg côtier sous la tutelle d’un maillage institutionnel prégnant, de la paroisse au quartier maritime. Puis, une première partie est consacrée à une étude de démographie historique, au moyen d’une analyse comparative entre le groupe socio-professionnel des gens de mer, qui représente plus de la moitié de la population, et les autres habitants. La reconstitution de plusieurs centaines de destinées a permis de dégager un comportement spécifique de la population maritime, quand il s’agit de convoler ou lorsqu’est venu le temps de rendre l’âme. Une deuxième partie est réservée à l’exploitation des ressources halieutiques et riveraines et à leur évolution, en analysant la façon dont les habitants tirent parti de cet espace convoité. En suivant le notaire lors des prisées, des indices du niveau et du mode de vie sont apparus après avoir ouvert les coffres et les armoires. Enfin, la dernière partie tend à souligner les particularités et les contraintes d’un environnement frontalier qui détermine la vie de tous les Saint-Vaastais, alors que les contours identitaires du groupe des gens de mer se dessinent au moyen de signifiants, dont le principal marqueur est constitué par un service obligatoire sur les vaisseaux de l’État royal, ce qui les différencie profondément du reste de la population. / This thesis is presented in the form of an investigation carried out in an 18th-century coastal society of Normandy, in order to apprehend the ties linking seafarers, people living by and of the sea, with the other inhabitants of the shore, as well as what may separate them, in a world that is, at first glance, turned towards the sea. The framework of the study is the natural harbour of La Hougue, on the east coast of the Cotentin, which shelters the population of Saint-Vaast-la-Hougue, the object of our investigations.An introductory chapter presents the conditions of opening onto the high sea off this coastal settlement under the supervision of a vivid institutional meshing, from the parish to the maritime district. Next, a first part is devoted to an historical study of demography, by means of a comparative analysis between the socio-professional group of seafarers, those living by and owing their livelihood to the sea and representing more than half the population, and the other inhabitants. The reconstruction of several hundred destinies has allowed for bringing out the specific behaviour of the maritime population when it comes to marriage or death. A second part is reserved for the exploitation of riparian and halieutic resources and their evolution, analysing the way in which the inhabitants put this coveted area to good use. By following the notary doing estimates, clues to living standards and the way of life came to light after safes and armoires were opened. Finally, the last part tends to emphasise the particularities and constraints of a border environment that determines the life of all the residents of Saint-Vaast, whereas the identity outlines of the sea folk become apparent by means of signifiers, of which the principal marker consisted of mandatory service on the vessels of the royal State, which differentiated them profoundly from the rest of the population.

No description available.