Puisque les traits de biométrie douce peuvent fournir des preuves supplémentaires pour aider à déterminer précisément l’identité de l’homme, il y a eu une attention croissante sur la reconnaissance faciale basée sur les biométrie douce ces dernières années. Parmi tous les biométries douces, le sexe et l’ethnicité sont les deux caractéristiques démographiques importantes pour les êtres humains et ils jouent un rôle très fondamental dans l’analyse de visage automatique. En attendant, la reconnaissance des expressions faciales est un autre challenge dans le domaine de l’analyse de visage en raison de la diversité et de l’hybridité des expressions humaines dans différentes cultures, genres et contextes. Ce thèse est dédié à combiner la texture du visage 2D et la morphologie du visage 3D pour estimer les biométries douces: le sexe, l’ethnicité, etc., et reconnaître les expressions faciales. Pour la reconnaissance du sexe et de l’ethnicité, nous présentons une approche efficace en combinant à la fois des textures locales et des caractéristiques de forme extraites à partir des modèles de visage 3D, contrairement aux méthodes existantes qui ne dépendent que des textures ou des caractéristiques de forme. Afin de souligne exhaustivement la différence entre les groupes sexuels et ethniques, nous proposons un nouveau descripteur, à savoir local circular patterns (LCP). Ce descripteur améliore Les motifs binaires locaux (LBP) et ses variantes en remplaçant la quantification binaire par une quantification basée sur le regroupement, entraînant d’une puissance plus discriminative et une meilleure résistance au bruit. En même temps, l’algorithme Adaboost est engagé à sélectionner les caractéristiques discriminatives fortement liés au sexe et à l’ethnicité. Les résultats expérimentaux obtenus sur les bases de données FRGC v2.0 et BU-3DFE démontrent clairement les avantages de la méthode proposée. Pour la reconnaissance des expressions faciales, nous présentons une méthode automatique basée sur les multi-modalité 2D + 3D et démontrons sa performance sur la base des données BU-3DFE. Notre méthode combine des textures locales et des descripteurs de formes pour atteindre l’efficacité et la robustesse. Tout d’abord, un grand ensemble des points des caractéristiques d’images 2D et de modèles 3D sont localisés à l’aide d’un nouvel algorithme, à savoir la cascade parallèle incrémentielle de régression linéaire (iPar-CLR). Ensuite, on utilise un nouveau descripteur basé sur les histogrammes des gradients d’ordre secondaire (HSOG) en conjonction avec le descripteur SIFT pour décrire la texture locale autour de chaque point de caractéristique 2D. De même, la géométrie locale autour de chaque point de caractéristique 3D est décrite par deux nouveaux descripteurs de forme construits à l’aide des quantités différentielle de géométries de la surface au premier ordre et au second ordre, à savoir meshHOG et meshHOS. Enfin, les résultats de reconnaissance des descripteurs 2D et 3D fournis par le classifier SVM sont fusionnés à la fois au niveau de fonctionnalité et de score pour améliorer la précision. Les expérimentaux résultats démontrent clairement qu’il existe des caractéristiques complémentaires entre les descripteurs 2D et 3D. Notre approche basée sur les multi-modalités surpasse les autres méthodes de l’état de l’art en obtenant une précision de reconnaissance 86,32%. De plus, une bonne capacité de généralisation est aussi présentée sur la base de données Bosphorus. / Since soft biometrics traits can provide sufficient evidence to precisely determine the identity of human, there has been increasing attention for face based soft biometrics identification in recent years. Among those face based soft biometrics, gender and ethnicity are both key demographic attributes of human beings and they play a very fundamental and important role in automatic machine based face analysis. Meanwhile, facial expression recognition is another challenge problem in face analysis because of the diversity and hybridity of human expressions among different subjects in different cultures, genders and contexts. This Ph.D thesis work is dedicated to combine 2D facial Texture and 3D face morphology for estimating people’s soft biometrics: gender, ethnicity, etc., and recognizing facial expression. For the gender and ethnicity recognition, we present an effective and efficient approach on this issue by combining both boosted local texture and shape features extracted from 3D face models, in contrast to the existing ones that only depend on either 2D texture or 3D shape of faces. In order to comprehensively represent the difference between different genders or ethnics groups, we propose a novel local descriptor, namely local circular patterns (LCP). LCP improves the widely utilized local binary patterns (LBP) and its variants by replacing the binary quantization with a clustering based one, resulting in higher discriminative power as well as better robustness to noise. Meanwhile, the following Adaboost based feature selection finds the most discriminative gender- and ethnic-related features and assigns them with different weights to highlight their importance in classification, which not only further raises the performance but reduces the time and memory cost as well. Experimental results achieved on the FRGC v2.0 and BU-3DFE data sets clearly demonstrate the advantages of the proposed method. For facial expression recognition, we present a fully automatic multi-modal 2D + 3D feature-based facial expression recognition approach and demonstrate its performance on the BU–3DFE database. Our approach combines multi-order gradientbased local texture and shape descriptors in order to achieve efficiency a nd robustness. First, a large set of fiducial facial landmarks of 2D face images along with their 3D face scans are localized using a novel algorithm namely incremental Parallel Cascade of Linear Regression (iPar–CLR). Then, a novel Histogram of Second Order Gradients (HSOG) based local image descriptor in conjunction with the widely used first-order gradient based SIFT descriptor are employed to describe the local texture around each 2D landmark. Similarly, the local geometry around each 3D landmark is described by two novel local shape descriptors constructed using the first-order and the second-order surface differential geometry quantities, i.e., Histogram of mesh Gradients (meshHOG) and Histogram of mesh Shape index (curvature quantization, meshHOS). Finally, the Support Vector Machine (SVM) based recognition results of all 2D and 3D descriptors are fused at both featurelevel and score-level to further improve the accuracy. Comprehensive experimental results demonstrate that there exist impressive complementary characteristics between the 2D and 3D descriptors. We use the BU–3DFE benchmark to compare our approach to the state-of-the-art ones. Our multi-modal feature-based approach outperforms the others by achieving an average recognition accuracy of 86,32%. Moreover, a good generalization ability is shown on the Bosphorus database.

Dans cette étude paléontologique, plusieurs méthodes ont été mises au point pour tester les hypothèses sur l’utilité des mustélidés en tant que signal paléoécologique, paléoenvironnemental et chronologique pour les sites préhistoriques du Paléolithique. Les genres Gulo, Martes, Mustela et Meles du Pléistocène et de l’Holocène de la moitié Sud de la France sont étudiés sous l’angle de l’anatomie comparée, de la biométrie et de la morphométrie géométrique sur un matériel très varié : crânien, dentaire et post-crânien. Les mustélidés proviennent de nombreux sites différents par leur âge, climat et altitude : Saint- Vallier, la Caune de l’Arago, Orgnac 3, le Lazaret, le Portel-Ouest, La Fage, La Marche, Siréjol, Villereversure, La Tanne du Beau Prince, la grotte des frères Traversat, Tuchan et Ambrussum. Cette approche systémique a permis de découvrir de nouveaux outils de sexage des mustélidés, sur les I3 notamment. L’indice de constriction orbitaire que nous avons établi livre une diagnose sexuelle très claire sur les blaireaux. La pente de la droite de régression des P4 et des Canines supérieures chez Mustela montre une distinction entre M. martes et M. foina. Nous confirmons la grande variabilité morphologique de ces petits Carnivores, qui reflète les environnements dans lesquels ils vivent ou ont vécu et qui est plus importante que la distinction spécifique chez Martes martes et Martes foina. De même, nous proposons Melesmeles atavus comme sous-espèce chronologique du Pléistocène moyen et présent à la Caune de l’Arago.De plus, les M. thorali, M. palerminea, M. praenivalis présentent plutôt des mandibules robustes et des dents graciles par rapport à leurs homologues actuels. / In this paleontological study, several methods have been developed to test hypotheses about the utility of mustelids as a paleoecological, palaeoenvironmental and chronological signal for prehistoric Paleolithic sites. The genera Gulo, Martes, Mustela and Meles of the Pleistocene and the Holocene of the half of the South of France are studied from the angle of comparative anatomy, biometry and geometrical morphometry on a very varied material: cranial, dental and postcranial. Mustelids come from various sites differed by age, climate and altitude: Saint-Vallier, Caune de l'Arago, Orgnac 3, Lazaret, Portel-Ouest, La Fage, La Marche, Siréjol, Villereversure, the Tanne du Beau Prince, Frères Traversat’s cave, Tuchan and Ambrussum. This systemic approach made it possible to discover new tools for sexing mustelids, particularly on I3. The post-orbital constriction index that we have established provides a very clear sexual diagnosis of badgers. The slope of the regression line of P4 and upper Canines of Mustela shows a distinction between M. martes and M. foina. We confirm the great morphological variability of these small carnivores, that reflects the environments in which they live or have lived and which is more important than the specific distinction in Martes martes and Martes foina. Similarly, we propose Meles meles atavus as a chronological subspecies of the Middle Pleistocene and present at the Caune de l'Arago. In addition, M. thorali, M. palerminea, M. praenivalis have strong mandibles and slender teeth compared to their current counterparts.

Le biome marin du Toarcien inférieur a subi une des crises les plus significatives de l'histoire de la Terre, qui a affecté les mers épicontinentales du globe. Ces bassins épicontinentaux sont également caractérisés par l'accumulation de grandes quantités de matière organique dans les sédiments de l'époque, ce qui a été interprété comme le résultat d'un Evénement Anoxique au Toarcien inférieur (T-OAE, ~183 Ma). Le nannoplancton calcaire, qui fait partie des organismes le plus sensibles aux conditions environnementales des eaux océaniques de surface, a été profondément affecté par cette crise et sa récupération se mit en place seulement à la fin du Toarcien inférieur. Si le T-OAE a été étudié en grand détail ces dernières années, la période de récupération est beaucoup moins bien connue. A ce stade, nous ne savons pas, notamment, si la récupération de la production carbonatée primaire est représentée par un épisode rapide ou par une suite d'événements (qui feraient partie d'un plus long processus) intervenus sur quelques millions d'années après le T-OAE. Pendant les perturbations environnementales qui ont eu lieu au Toarcien inférieur, des fluctuations importantes du niveau marin ont été interprétées ainsi que des variations de la température des eaux océaniques. Après le T-OAE, le niveau marin et les températures montrent des changements plus graduels pendant une grande partie du Toarcien supérieur-Aalénien inférieur et les niches écologiques occupées par le nannoplancton calcaire sont ré-établies. Ces conditions de relative stabilité sur le long terme (~10 Ma) sont idéales pour étudier les dynamiques de la communauté des nannofossiles calcaires et, notamment, comment les différents taxons réorganisent leurs relations synécologiques. Le Toarcien-Aalénien est une période clé dans l'évolution du nannoplancton calcaire, avec des nouveaux genres qui apparaissent. Les nannofossiles de cet intervalle restent relativement peu étudié dans le Bassin Lusitanien, malgré l'importance de cette région d'un point de vue paléocéanographique. En effet ce bassin, qui était à l'origine un rift avorté lié à l'ouverture de l'Atlantique, a permis les connections entre des masses d'eaux de provenance NW européenne et sud-téthysienne. C'est dans cette région que nous retrouvons donc des mélanges de taxons liés à des différentes provinces géographiques / At ~183 Ma when large amounts of organic matter was dumped onto the ocean beds (known as the Toarcian Oceanic Anoxic Event, T-OAE), the Early Toarcian marine biota experienced one of the most important biological crises in Earth history. Calcareous nannoplankton was deeply affected and started only to recover from the end of the Early Toarcian. After the T-OAE, as phytoplankton communities dwelling the oceans photic layer recover from the previous disturbing conditions, the ecological niches once occupied by calcareous nannoplankton are again replenished. Located in the westernmost part of the Tethyan ocean, the Lusitanian Basin acted as a seaway between the NW Tethys and the Mediterranean Tethys provinces, therefore been actively influenced by these two different water masses. Hence such a region stands as the perfect spot to study the calcareous nannoplankton community, as its fossil remains bears species typical of both provinces. In order to appraise the nannoplankton demise and subsequent recovery, absolute and relative abundances of nannofossils were determined from pelagic marl-limestone couplets from three sections of the Lusitanian Basin. Also brachiopods shells were investigated for stable carbon and oxygen isotopes analysis. In order to independently assess the morphologic evolution of Lotharingius coccoliths, a Toarcian section from south France was studied. Different statistic methodologies were used throughout this work, in order to screen for robust and significant information from the numerous datasets that were built / Há cerca de 183 milhões de anos atrás, a biota marinha sofreu uma das mais importantes crises biológicas na história do planeta, quando grandes quantidades de matéria orgânica foram depositadas nos fundos oceânicos. Este episódio que teve lugar durante o Toarciano, resultou de um Evento Anóxico Oceânico (T-OAE), e afectou os mares epicontinentais à escala global. O nanoplâncton calcário, um grupo de organismos extremamente sensível às variações nas condições ambientais que têm lugar na camada superficial das massas de água, foi profundamente afectado. Foi somente no final do Toarciano Inferior que a sua recuperação se iniciou. Apesar de já muito se ter escrito sobre o T-OAE, a previsível recuperação deste grupo nunca foi sistemática nem detalhadamente estudada. Até hoje, não se sabia ainda se esta recuperação teria sido relativamente rápida, ou resultado de um processo lento e gradual que teria durado milhões de anos. Durante o Toarciano Inferior, quando as condições ambientais eram extremamente instáveis, ocorreram importantes e drásticas oscilações na temperatura e no nível médio dos oceanos. Após o T-OAE e durante todo o Toarciano e Aaleniano Inferior, as oscilações do nível médio dos oceanos e de temperatura são bastante mais suaves. À medida que a comunidade fitoplanctónica instalada na zona fótica recupera das condições extremas anteriores, os nichos ecológicos antes ocupados por nanoplâncton calcário, começam novamente a ser preenchidos. Apesar das prolongadas e suaves variações abióticas, é num ambiente então mais estável, que melhor se pode observar a evolução deste grupo, e de que forma as suas relações sinecológicas se alteraram ao longo de aproximadamente 10 milhões de anos. Apesar de ser considerado como um período onde a comunidade nanoplanctónica sofreu importantes alterações, nada se sabe em concreto que alterações foram essas que tiveram lugar após o TOAE num local tão estratégico como a Bacia Lusitânica. Situada na extremidade Oeste do Tétis e fazendo parte de um rift originado pela abertura do Atlântico, a Bacia Lusitânica actuou como um corredor que ligava livremente as massas de água das províncias mediterrânica e NW do Tétis, sendo por estas, activamente influenciada. Por essa razão, tal situação geográfica constitui um local excepcional para se estudar a comunidade de nanoplâncton calcário, uma vez que o seu registo fóssil compreende espécies típicas destas duas províncias do Tétis

A Computação Ciente de Contexto permite a obtenção e utilização de informações de contexto adquiridas de dispositivos computacionais no ambiente, com o objetivo de prover serviços; esta dinâmica aliada à evolução das redes de computadores vem provocando profundas modificações nos aspectos sociais e comportamentais das pessoas, uma vez que gradativamente têm necessidade de viverem imersas na tecnologia e integradas ao ambiente, com transparência e mobilidade, e de tal forma que as aplicações de software se adaptam ao comportamento das pessoas e nas informações de contexto capturadas do ambiente. Um dos desafios desta interação ser humano - ambiente - tecnologia - ubiqüidade é garantir a segurança. Como principal inovação e contribuição, esta tese propõe um mecanismo de autenticação contínua de usuários que faz uso de informações de contexto do ambiente, da análise do comportamento do usuário, da biometria facial, das teorias comportamentais de Skinner e da Confiança Matemática da Teoria das Evidências de Dempster-Shafer, para compor uma política de segurança adaptativa e um Sistema de Autenticação Contínua de Usuários Conhecidos - KUCAS (Known User Continuous Authentication System), que estabelece níveis de confiança para autenticar o usuário através da análise do comportamento dele em um ambiente ou domínio específico nas redes de computadores, num determinado período de tempo. A dinâmica de gerenciamento incluso nesse sistema compara o comportamento atual com o histórico de comportamentos anteriores do usuário e com as restrições de atribuição de confiança; caso haja indícios de mudanças no comportamento do usuário, aciona por meio de sensores, a Tecnologia de Reconhecimento Facial Tridimensional (3D), que captura a imagem da face do usuário, validando-a e armazenando-a nos bancos de dados de imagens; havendo incertezas e divergências, mecanismos de segurança e sinais de alerta são acionados. O Sistema KUCAS proposto possui uma infra-estrutura de um framework F-KUCAS, um Módulo de Segurança S-KUCAS e um Algoritmo de Autenticação A-KUCAS. / Context-aware Computing allows to obtain and use context informations acquired through devices in the environment, with the goal to provide services. This dynamics, allied to the computer networks evolution, has been provoking deep modifications in peoples social and behavior aspects, seeing that they have the necessity to live immersed in technology and integrated with the environment, with transparency and mobility, anywhere, anytime, so that the software applications adapt themselves to the persons behavior, based on the context information captured through the environment. One of the challenges of this human ? environment - technology ? ubiquity interaction is to provide security. As main innovation and contribution, this thesis presents an authentication mechanism of users which makes use of environmental context information, users behavior analysis, the face recognition technology, the behavior theories of Skinner and the Mathematical Confidence of the Theory of the Evidences of Dempster-Shafer, to compose an adaptative security policy and the Known User Continuous Authentication System (KUCAS) that establishes trust levels to authenticate the user by his behavior analysis in a specific domain of the computer networks, in a period of time. The dynamics of enclosed management in this system compares the current behavior with the users previous behaviors description and with the trust restrictions. In case of indications of changes in the users behavior, the 3D Technology Face Recognition is set in motion by sensors, which capture the image of the users face, validating it and storing it in the data bases of images. If there are uncertainties and divergences, mechanisms of security and signals of alert are set in motion. The KUCAS System has an infrastructure of one framework F-KUCAS, a Security Module S-KUCAS and an Algorithm of Authentication A-KUCAS.

The Federal Bureau of Investigation reported that cyber actors will likely increase cyber intrusions against healthcare systems and their concomitant medical devices because of the mandatory transition from paper to electronic health records, lax cyber security standards, and a higher financial payout for medical records in the deep web. The problem addressed in this quantitative correlational study was uncertainty surrounding the benefits of palm vein authentication adoption relative to the growing crime of medical identity theft. The purpose of this quantitative correlational study was to understand healthcare managers' and doctors' perceptions of the effectiveness of palm vein authentication technology. The research questions were designed to investigate the relationship between intention to adopt palm vein authentication technology and perceived usefulness, complexity, security, peer influence, and relative advantage. The unified theory of acceptance and use of technology was the theoretical basis for this quantitative study. Data were gathered through an anonymous online survey of 109 healthcare managers and doctors, and analyzed using principal axis factoring, Pearson's product moment correlation, multiple linear regression, and 1-way analysis of variance. The results of the study showed a statistically significant positive correlation between perceived usefulness, security, peer influence, relative advantage, and intention to adopt palm vein authentication. No statistically significant correlation existed between complexity and intention to adopt palm vein authentication. These findings indicate that by effectively using palm vein authentication, organizations can mitigate the risk of medical fraud and its associated costs, and positive social change can be realized.

For this project, we use a modified version of a previously developed mathematical model, which describes the relationships among matrix metalloproteinases (MMPs), their tissue inhibitors (TIMPs), and extracellular matrix (ECM). Our ultimate goal is to quantify and understand differences in parameter estimates between patients in order to predict future responses and individualize treatment for each patient. By analyzing parameter confidence intervals and confidence and prediction intervals for the state variables, we develop a parameter space reduction algorithm that results in better future response predictions for each individual patient. Moreover, use of another subset selection method, namely Structured Covariance Analysis, that considers identifiability of parameters, has been included in this work. Furthermore, to estimate parameters more efficiently and accurately, the standard error (SE- )optimal design method is employed, which calculates optimal observation times for clinical data to be collected. Finally, by combining different parameter subset selection methods and an optimal design problem, different cases for both finding optimal time points and intervals have been investigated.

Blood coagulation is a series of biochemical reactions that take place to form a blood clot. Abnormalities in coagulation, such as under-clotting or over- clotting, can lead to significant blood loss, cardiac arrest, damage to vital organs, or even death. Thus, understanding quantitatively how blood coagulation works is important in informing clinical decisions about treating deficiencies and disorders. Quantifying blood coagulation is possible through mathematical modeling. This review presents different mathematical models that have been developed in the past 30 years to describe the biochemistry, biophysics, and clinical applications of blood coagulation research. This review includes the strengths and limitations of models, as well as suggestions for future work.

<p>This thesis states that cracks and scars, referred to as Secondary Creases, in fingerprint images can be used as means for aiding and complementing fingerprint recognition, especially in cases where there is not enough clear data to use traditional methods such as minutiae based or correlation techniques. A Gabor filter bank is used to extract areas with linear patterns, where after the Hough Transform is used to identify secondary creases in a r, theta space. The methods proposed for Secondary Crease extraction works well, and provides information about what areas in an image contains usable linear pattern. Methods for comparison is however not as robust, and generates False Rejection Rate at 30% and False Acceptance Rate at 20% on the proposed dataset that consists of bad quality fingerprints. In short, our methods still makes it possible to make use of fingerprint images earlier considered unusable in fingerprint recognition systems.</p>

Automatic verification of personal identity using facial images is the central topic of the thesis. This problem can be stated as follows. Given two face images, it must be determined automatically whether they are images of the same person or of different persons. Due to many factors such as variability of facial appearance, sensitivity to noise, template aging, etc., the problem is difficult. We can overcome some of these difficulties by combining different information sources for the classification/recognition task. In this thesis we propose strategies on how to combine the different information sources, i.e. fusion strategies, in order to improve the verification accuracy. We have designed and thoroughly optimised a number of face verification algorithms. Their individual properties such as how their accuracy depends on algorithm parameters, image size, or sensitivity to mis-registrations have been studied. We have also studied how to combine the outputs of the different algorithms in order to reduce the verification error rates. Another decision fusion aspect considered in this thesis is the fusion of confidences obtained sequentially on several video frames of the same person's face. Finally multimodal fusion has been studied. In this case, the speech and face of the same subject are recorded and processed by different algorithms which output separate opinions. These two opinions are then conciliated at the fusion stage. It is shown that in all cases, information fusion allows a considerable performance improvement if the fusion stage is carefully designed.

Biometric systems allow identification of human persons based on physiological or behavioral characteristics, such as voice, handprint, iris or facial characteristics. The use of face and iris recognition as a way to authenticate user’s identities has been a topic of research for years. Present iris recognition systems require that subjects stand close (<2m) to the imaging camera and look for a period of about three seconds until the data are captured. This cooperative behavior is required in order to capture quality images for accurate recognition. This will eventually restrict the amount of practical applications where iris recognition can be applied, especially in an uncontrolled environment where subjects are not expected to cooperate such as criminals and terrorists, for example. For this reason, this thesis develops a collection of methods to deal with low quality face and iris images and that can be applied for face and iris recognition in a non-cooperative environment. This thesis makes the following main contributions: I. For eye and face tracking in low quality images, a new robust method is developed. The proposed system consists of three parts: face localization, eye detection and eye tracking. This is accomplished using traditional image-based passive techniques such as shape information of the eye and active based methods which exploit the spectral properties of the pupil under IR illumination. The developed method is also tested on underexposed images where the subject shows large head movements. II. For iris recognition, a new technique is developed for accurate iris segmentation in low quality images where a major portion of the iris is occluded. Most existing methods perform generally quite well but tend to overestimate the occluded regions, and thus lose iris information that could be used for identification. This information loss is potentially important in the covert surveillance applications we consider in this thesis. Once the iris region is properly segmented using the developed method, the biometric feature information is calculated for the iris region using the relative entropy technique. Iris biometric feature information is calculated using two different feature decomposition algorithms based on Principal Component Analysis (PCA) and Independent Component Analysis (ICA). III. For face recognition, a new approach is developed to measure biometric feature information and the changes in biometric sample quality resulting from image degradations. A definition of biometric feature information is introduced and an algorithm to measure it proposed, based on a set of population and individual biometric features, as measured by a biometric algorithm under test. Examples of its application were shown for two different face recognition algorithms based on PCA (Eigenface) and Fisher Linear Discriminant (FLD) feature decompositions.