La robotique mobile est un domaine en plein essor. L'un des domaines de recherche consiste à permettre à un robot de cartographier son environnement tout en se localisant dans l'espace. Les techniques couramment employées de SLAM (Simultaneous Localization And Mapping) restent généralement coûteuses en termes de puissance de calcul. La tendance actuelle vers la miniaturisation des systèmes impose de restreindre les ressources embarquées. L'ensemble de ces constatations nous ont guidés vers l'intégration d'algorithmes de SLAM sur des architectures adéquates dédiées pour l’embarqué.Les premiers travaux ont consisté à définir une architecture permettant à un robot mobile de se localiser. Cette architecture doit respecter certaines contraintes, notamment celle du temps réel, des dimensions réduites et de la faible consommation énergétique.L’implantation optimisée d’un algorithme (EKF-SLAM), en utilisant au mieux les spécificités architecturales du système (capacités des processeurs, implantation multi-cœurs, calcul vectoriel ou parallélisation sur architecture hétérogène), a permis de démontrer la possibilité de concevoir des systèmes embarqués pour les applications SLAM dans un contexte d’adéquation algorithme architecture. Une seconde approche a été explorée ayant pour objectif la définition d’un système à base d’une architecture reconfigurable (à base de FPGA) permettant la conception d'une architecture fortement parallèle dédiée au SLAM. L'architecture définie a été évaluée en utilisant une méthodologie HIL (Hardware in the Loop).Les principaux algorithmes de SLAM sont conçus autour de la théorie des probabilités, ils ne garantissent en aucun cas les résultats de localisation. Un algorithme de SLAM basé sur la théorie ensembliste a été défini garantissant l'ensemble des résultats obtenus. Plusieurs améliorations algorithmiques sont ensuite proposées. Une comparaison avec les algorithmes probabilistes a mis en avant la robustesse de l’approche ensembliste.Ces travaux de thèse mettent en avant deux contributions principales. La première consiste à affirmer l'importance d'une conception algorithme-architecture pour résoudre la problématique du SLAM. La seconde est la définition d’une méthode ensembliste permettant de garantir les résultats de localisation et de cartographie. / Mobile robotics is a growing field. One important research area is Simultaneous Localization And Mapping (SLAM). Algorithms commonly used in SLAM are generally expensive in terms of computing power. The current trend towards miniaturization imposes to restrict the embedded processing units. All these observations lead us to integrate SLAM algorithms on dedicated architectures suitable for embedded systems.The first work was to define an architecture for a mobile robot to localize itself. This architecture must satisfy some constraints, including the real-time, small dimensions and low power consumption. The optimized implementation of a SLAM algorithm, using the best architectural characteristics of the system (capacity of processors, multi-core implementation, SIMD instructions or parallelization on heterogeneous architecture), has demonstrated the ability to design embedded systems for SLAM applications in the context of Hardware-Software codesign.A second approach has been explored with the aim of designing a system based on a reconfigurable architecture (FPGA-based) for a highly parallel architecture dedicated to SLAM. The defined architecture was evaluated using a HIL (Hardware in the Loop) methodology.The main SLAM algorithms use the probabilistic theories, they do not guarantee their localization results. A SLAM algorithm based on interval analysis is defined to guarantee the overall results. Several algorithmic improvements are then proposed. A comparison with probabilistic algorithms highlighted the robustness of the approach.This thesis put forward two main contributions. The first is to affirm the importance of the hardware software codesign to solve the problem of SLAM with real-time constraint. The second is the definition of a new algorithm to ensure the results of localization and mapping.

Au cours de cette thèse, nous nous sommes intéressés à deux principales problématiques de la télédétection spatiale de milieux urbains qui sont : le "démélange spectral " et la "fusion". Dans la première partie de la thèse, nous avons étudié le démélange spectral d'images hyperspectrales de scènes de milieux urbains. Les méthodes développées ont pour objectif d'extraire, d'une manière non-supervisée, les spectres des matériaux présents dans la scène imagée. Le plus souvent, les méthodes de démélange spectral (méthodes dites de séparation aveugle de sources) sont basées sur le modèle de mélange linéaire. Cependant, lorsque nous sommes en présence de paysage non-plat, comme c'est le cas en milieu urbain, le modèle de mélange linéaire n'est plus valide et doit être remplacé par un modèle de mélange non-linéaire. Ce modèle non-linéaire peut être réduit à un modèle de mélange linéaire-quadratique/bilinéaire. Les méthodes de démélange spectral proposées sont basées sur la factorisation matricielle avec contrainte de non-négativité, et elles sont conçues pour le cas particulier de scènes urbaines. Les méthodes proposées donnent généralement de meilleures performances que les méthodes testées de la littérature. La seconde partie de cette thèse à été consacrée à la mise en place de méthodes qui permettent la fusion des images multispectrale et hyperspectrale, afin d'améliorer la résolution spatiale de l'image hyperspectrale. Cette fusion consiste à combiner la résolution spatiale élevée des images multispectrales et la haute résolution spectrale des images hyperspectrales. Les méthodes mises en place sont des méthodes conçues pour le cas particulier de fusion de données de télédétection de milieux urbains. Ces méthodes sont basées sur des techniques de démélange spectral linéaire-quadratique et utilisent la factorisation en matrices non-négatives. Les résultats obtenus montrent que les méthodes développées donnent globalement des performances satisfaisantes pour la fusion des données hyperspectrale et multispectrale. Ils prouvent également que ces méthodes surpassent significativement les approches testées de la littérature. / In this thesis, we focused on two main problems of the spatial remote sensing of urban environments which are: "spectral unmixing" and "fusion". In the first part of the thesis, we are interested in the spectral unmixing of hyperspectral images of urban scenes. The developed methods are designed to unsupervisely extract the spectra of materials contained in an imaged scene. Most often, spectral unmixing methods (methods known as blind source separation) are based on the linear mixing model. However, when facing non-flat landscape, as in the case of urban areas, the linear mixing model is not valid any more, and must be replaced by a nonlinear mixing model. This nonlinear model can be reduced to a linear-quadratic/bilinear mixing model. The proposed spectral unmixing methods are based on matrix factorization with non-negativity constraint, and are designed for urban scenes. The proposed methods generally give better performance than the tested literature methods. The second part of this thesis is devoted to the implementation of methods that allow the fusion of multispectral and hyperspectral images, in order to improve the spatial resolution of the hyperspectral image. This fusion consists in combining the high spatial resolution of multispectral images and high spectral resolution of hyperspectral images. The implemented methods are designed for urban remote sensing data. These methods are based on linear-quadratic spectral unmixing techniques and use the non-negative matrix factorization. The obtained results show that the developed methods give good performance for hyperspectral and multispectral data fusion. They also show that these methods significantly outperform the tested literature approaches.

Depuis l'invention de l'automobile à la fin du 19eme siècle, la taille du parc ainsi que l'importance du trafic routier n'ont cessées d'augmenter. Ceci a malheureusement été suivi par l'augmentation constante du Nombre d'accidents routiers. Un grand nombre d'études et notamment un rapport fourni par l'Organisation Mondiale de la Santé, a présenté un état alarmant du nombre de blessés et de décès liés aux accidents routiers. Afin de réduire ces chiffres, une solution réside dans le Développement de systèmes d'aide à la conduite qui ont pour but d'assister le conducteur dans sa tâche de conduite. La recherche dans le domaine des aides à la conduite s'est montrée très dynamique et productive durant les vingt dernières années puisque des systèmes tels que l'antiblocage de sécurité (ABS), le programme de stabilité électronique (ESP), le régulateur de vitesse intelligent (ACC), l'assistant aux manœuvres de parking (PMA), les phares orientables (DBL), etc. sont maintenant commercialisés et acceptés par la majorité des conducteurs. Cependant, si ces systèmes ont permis d'améliorer la sécurité des conducteurs, de nombreuses pistes sont encore à explorer. En effet, les systèmes d'aide à la conduite existants ont un comportement microscopique, en d'autres termes ils se focalisent uniquement sur la tâche qu'ils ont à effectuer. Partant du principe que la collaboration entre toutes ces aides à la conduite est plus efficace que leur utilisation en parallèle, une approche globale d'aide à la conduite devient nécessaire. Ceci se traduit par la nécessité de développer une nouvelle génération d'aide à la conduite, prenant en compte d'avantage d'informations et de contraintes liées au véhicule, au conducteur et à son environnement. [...] / Since the origin of the automotive at the end of the 19th century, the traffic flow is subject to a constant increase and, unfortunately, involves a constant augmentation of road accidents. Research studies such as the one performed by the World Health Organization, show alarming results about the number of injuries and fatalities due to these accidents. To reduce these figures, a solution lies in the development of Advanced Driver Assistance Systems (ADAS) which purpose is to help the Driver in his driving task. This research topic has been shown to be very dynamic and productive during the last decades. Indeed, several systems such as Anti-lock Braking System (ABS), Electronic Stability Program (ESP), Adaptive Cruise Control (ACC), Parking Manoeuvre Assistant (PMA), Dynamic Bending Light (DBL), etc. are yet market available and their benefits are now recognized by most of the drivers. This first generation of ADAS are usually designed to perform a specific task in the Controller/Vehicle/Environment framework and thus requires only microscopic information, so requires sensors which are only giving local information about an element of the Vehicle or of its Environment. On the opposite, the next ADAS generation will have to consider more aspects, i.e. information and constraints about of the Vehicle and its Environment. Indeed, as they are designed to perform more complex tasks, they need a global view about the road context and the Vehicle configuration. For example, longitudinal control requires information about the road configuration (straight line, bend, etc.) and about the eventual presence of other road users (vehicles, trucks, etc.) to determine the best reference speed. [...]

Aujourd'hui c'est un fait, nous vivons dans un monde où les enjeux écologiques, économiques et sociétaux sont de plus en plus pressants. Au croisement des différentes lignes directrices envisagées pour répondre à ces problèmes, une vision plus précise de la mobilité humaine est un axe central et majeur, qui a des répercussions sur tous les domaines associés tels que le transport, les sciences sociales, l'urbanisme, les politiques d'aménagement, l'écologie, etc. C'est par ailleurs dans un contexte de contraintes budgétaires fortes que les principaux acteurs de la mobilité sur les territoires cherchent à rationaliser les services de transport, et les déplacements des individus. La mobilité humaine est donc un enjeu stratégique aussi bien pour les collectivités locales que pour les usagers, qu'il faut savoir observer, comprendre, et anticiper.Cette étude de la mobilité passe avant tout par une observation précise des déplacements des usagers sur les territoires. Aujourd'hui les acteurs de la mobilité se tournent principalement vers l'utilisation massive des données utilisateurs. L'utilisation simultanée de données multi-sources, multi-modales, et multi-échelles permet d'entrevoir de nombreuses possibilités, mais cette dernière présente des défis technologiques et scientifiques majeurs. Les modèles de mobilité présentés dans la littérature sont ainsi trop souvent axés sur des zones d'expérimentation limitées, en utilisant des données calibrées, etc. et leur application dans des contextes réels, et à plus large échelle est donc discutable. Nous identifions ainsi deux problématiques majeures qui permettent de répondre à ce besoin d'une meilleure connaissance de la mobilité humaine, mais également à une meilleure application de cette connaissance. La première problématique concerne l'extraction d'informations de mobilité à partir de la fusion de données hétérogènes. La seconde problématique concerne la pertinence de cette fusion dans un contexte réel, et à plus large échelle. Nous apportons différents éléments de réponses à ces problématiques dans cette thèse. Tout d'abord en présentant deux modèles de fusion de données, qui permettent une extraction d'informations pertinentes. Puis, en analysant l'application de ces deux modèles au sein du projet ANR Norm-Atis.Dans cette thèse, nous suivons finalement le développement de toute une chaine de processus. En commençant par une étude de la mobilité humaine, puis des modèles de mobilité, nous présentons deux modèles de fusion de données, et nous analysons leur pertinence dans un cas concret. Le premier modèle que nous proposons permet d'extraire 12 comportements types de mobilité. Il est basé sur un apprentissage non-supervisé de données issues de la téléphonie mobile. Nous validons nos résultats en utilisant des données officielles de l'INSEE, et nous déduisons de nos résultats, des comportements dynamiques qui ne peuvent pas être observés par les données de mobilité traditionnelles. Ce qui est une forte valeur-ajoutée de notre modèle. Le second modèle que nous proposons permet une désagrégation des flux de mobilité en six motifs de mobilité. Il se base sur un apprentissage supervisé des données issues d'enquêtes de déplacements ainsi que des données statiques de description du sursol. Ce modèle est appliqué par la suite aux données agrégés au sein du projet Norm-Atis. Les temps de calculs sont suffisamment performants pour permettre une application de ce modèle dans un contexte temps-réel. / Today it is a fact that we live in a world where ecological, economic and societal issues are increasingly pressing. At the crossroads of the various guidelines envisaged to address these problems, a more accurate vision of human mobility is a central and major axis, which has repercussions on all related fields such as transport, social sciences, urban planning, management policies, ecology, etc. It is also in the context of strong budgetary constraints that the main actors of mobility on the territories seek to rationalize the transport services and the movements of individuals. Human mobility is therefore a strategic challenge both for local communities and for users, which must be observed, understood and anticipated.This study of mobility is based above all on a precise observation of the movements of users on the territories. Nowadays mobility operators are mainly focusing on the massive use of user data. The simultaneous use of multi-source, multi-modal, and multi-scale data opens many possibilities, but the latter presents major technological and scientific challenges. The mobility models presented in the literature are too often focused on limited experimental areas, using calibrated data, etc., and their application in real contexts and on a larger scale is therefore questionable. We thus identify two major issues that enable us to meet this need for a better knowledge of human mobility, but also to a better application of this knowledge. The first issue concerns the extraction of mobility information from heterogeneous data fusion. The second problem concerns the relevance of this fusion in a real context, and on a larger scale. These issues are addressed in this dissertation: the first, through two data fusion models that allow the extraction of mobility information, the second through the application of these fusion models within the ANR Norm-Atis project.In this thesis, we finally follow the development of a whole chain of processes. Starting with a study of human mobility, and then mobility models, we present two data fusion models, and we analyze their relevance in a concrete case. The first model we propose allows to extract 12 types of mobility behaviors. It is based on an unsupervised learning of mobile phone data. We validate our results using official data from the INSEE, and we infer from our results, dynamic behaviors that can not be observed through traditional mobility data. This is a strong added-value of our model. The second model operates a mobility flows decompositoin into six mobility purposes. It is based on a supervised learning of mobility surveys data and static data from the land use. This model is then applied to the aggregated data within the Norm-Atis project. The computing times are sufficiently powerful to allow an application of this model in a real-time context.

Les maladies chroniques impliquant le système respiratoire nécessitent un suivi sur la durée. L’activité physique et les paramètres cardiovasculaires sont essentiels pour ces pathologies. Nous nous sommes intéressés en particulier à la BPCO et à l’apnée obstructive du sommeil. La BPCO est caractérisée par un cercle vicieux d’inactivité : une gêne respiratoire entraîne une diminution de l’activité, qui elle-même augmente la gêne respiratoire par désentraînement. Le monitoring de l’activité, en lien avec la SpO2 est donc essentiel pour cette pathologie. Les désaturations nocturnes sont un paramètre cardinal de l’apnée du sommeil. Un actimètre permet d’évaluer la qualité du sommeil, complétant ainsi le suivi de cette pathologie. De plus, l’activité diurne est un indicateur de l’asthénie provoquée par le syndrome. Le but de ce travail a donc été la mise au point d’un actimètre communicant, capable de mesurer l’activité diurne, d’évaluer le temps de sommeil et de s’interfacer avec un oxymètre de pouls pour synchroniser la collecte de données. À partir des données récoltées durant 26 jours d’enRégistrements, nous avons mis au point et évalué un algorithme permettant de mesurer le temps passé assis, debout et allongé. Cet algorithme a été conçu pour être embarqué dans un microcontrôleur, ayant des ressources de calcul limitées. Nous avons également proposé un algorithme de détection des pas, dont le fonctionnement a été validé sur plus de 5 heures de marche, sur 22 patients différents, contre un comptage manuel. Nous avons enfin proposé une méthode de détection des transitions assis-debout pour l’instrumentation du test de levers de chaise de 3 minutes. Lors de l’analyse nocturne, nous avons mis au point un algorithme de détection du temps de sommeil, testé sur 25 nuits. Nous avons également proposé une méthode d’analyse de l’onde de pouls permettant d’extraire le rapport LF/HF de la variabilité cardiaque, permettant de détecter le sommeil paradoxal. Nous avons montré le résultat de l’agrégation des différentes données acquises par le système formé de l’actimètre et de l’oxymètre lors d’une nuit d’examen, comme outils à disposition du praticien. L’actimètre mis au point dans le cadre de ces travaux et les méthodes d’analyse du signal associées sont adaptés au suivi non invasif de pathologies respiratoires. Ils peuvent également être intégrés à un système de télémédecine via une passerelle informatique pour un suivi de long terme. / Chronic diseases affecting the respiratory system require a long-term monitoring. Physical activity and cardiovascular parameters are essential in those pathologies. We focused on two of those diseases : COPD and obstructive sleep apnea. COPD is characterized by a downward cycle of inactivity : a respiratory impairment leads to a reduction of activity, whose in turn worsen the respiratory impairment by a conditioning loss. As a consequence, activity monitoring and SpO2 are essential for the monitoring of this pathology. Nocturnal oxygen desaturation are a main feature of sleep apnea. An actimeter allows for sleep quality evaluation, and is a logical choice for a complementary measure of this disease. Moreover, diurnal activity is an indicator of the degree of physical weakness that can occur as a consequence of sleep apnea. The main goal of the work has been the developement of a connected actimeter, able to monitor diurnal activity, estimate the duration of sleep and collect data from a pulse oximeter to synchronise the data. From 26 days of accelerometric measures, we designed and validated an algorithm that compute the time spend sitting, standing and lying. This algorithm has been designed to be embedded in a microcontroler with limited computing power. We also proposed a step detection algorithm validated on 5 hours of walking, on 22 different patients, against a visual count. Finally, we designed a method to detect the sitting-standing change of posture to monitor the 3-minutes chair stand test. On the nocturnal aspect, we designed an algorithm used to estimate the sleep duration during a night. It as been tested on 25 nights. We also proposed a pulse wave analysis method to extract the LF/HF ratio of cardiac variability, to detect REM sleep. We showed the result of the aggregation of the different parameters collected by the system composed of the actimeter and the oximeter during a monitored night, as a tool to the healthcare professional. The actimeter design in the context of this work and the associated signal processing methods are appropriate to the monitoring of respiratory pathologies with a light equipment. They also can be integrated into a telemedecine system through a gateway computer, allowing for a long-term monitoring.

The purpose of information acquisition for data and information fusion is to provide relevant and timely information. The acquired information is integrated (or fused) to estimate the state of some environment. The success of information acquisition can be measured in the quality of the environment state estimates generated by the data and information fusion process. In this thesis, we introduce and set out to characterise the concept of large-scale information acquisition. Our interest in this subject is justified both by the identified lack of research on a holistic view on data and information fusion, and the proliferation of networked sensors which promises to enable handy access to a multitude of information sources. We identify a number of properties that could be considered in the context of large-scale information acquisition. The sensors used could be large in number, heterogeneous, complex, and distributed. Also, algorithms for large-scale information acquisition may have to deal with decentralised control and multiple and varying objectives. In the literature, a process that realises information acquisition is frequently denoted sensor management. We, however, introduce the term perception management instead, which encourages an agent perspective on information acquisition. Apart from explictly inviting the wealth of agent theory research into the data and information fusion research, it also highlights that the resource usage of perception management is constrained by the overall control of a system that uses data and information fusion. To address the challenges posed by the concept of large-scale information acquisition, we present a framework which highlights some of its pertinent aspects. We have implemented some important parts of the framework. What becomes evident in our study is the innate complexity of information acquisition for data and information fusion, which suggests approximative solutions. We, furthermore, study one of the possibly most important properties of large-scale information acquisition, decentralised control, in more detail. We propose a recurrent negotiation protocol for (decentralised) multi-agent coordination. Our approach to the negotiations is from an axiomatic bargaining theory perspective; an economics discipline. We identify shortcomings of the most commonly applied bargaining solution and demonstrate in simulations a problem instance where it is inferior to an alternative solution. However, we can not conclude that one of the solutions dominates the other in general. They are both preferable in different situations. We have also implemented the recurrent negotiation protocol on a group of mobile robots. We note some subtle difficulties with transferring bargaining solutions from economics to our computational problem. For instance, the characterising axioms of solutions in bargaining theory are useful to qualitatively compare different solutions, but care has to be taken when translating the solution to algorithms in computer science as some properties might be undesirable, unimportant or risk being lost in the translation. / QC 20100903

This dissertation is concerned with the problem of describing the singing voice within the audio signal of a song. This work is motivated by the fact that the lead vocal is the element that attracts the attention of most listeners. For this reason it is common for music listeners to organize and browse music collections using information related to the singing voice such as the singer name. Our research concentrates on the three major problems of music information retrieval: the localization of the source to be described (i.e. the recognition of the elements corresponding to the singing voice in the signal of a mixture of instruments), the search of pertinent features to describe the singing voice, and finally the development of pattern recognition methods based on these features to identify the singer. For this purpose we propose a set of novel features computed on the temporal variations of the fundamental frequency of the sung melody. These features, which aim to describe the vibrato and the portamento, are obtained with the aid of a dedicated model. In practice, these features are computed on the time-varying frequency of partials obtained using the sinusoidal model. In the first experiment we show that partials corresponding to the singing voice can be accurately differentiated from the partials produced by other instruments using decisions based on the parameters of the vibrato and the portamento. Once the partials emitted by the singer are identified, the segments of the song containing singing can be directly localized. To improve the recognition of the partials emitted by the singer we propose to group partials that are related harmonically. Partials are clustered according to their degree of similarity. This similarity is computed using a set of CASA cues including their temporal frequency variations (i.e. the vibrato and the portamento). The clusters of harmonically related partials corresponding to the singing voice are identified using the vocal vibrato and the portamento parameters. Groups of vocal partials can then be re-synthesized to isolate the voice. The result of the partial grouping can also be used to transcribe the sung melody. We then propose to go further with these features and study if the vibrato and portamento characteristics can be considered as a part of the singers' signature. Previous works on singer identification describe audio signals using features extracted on the short-term amplitude spectrum. The latter features aim to characterize the timbre of the sound, which, in the case of singing, is related to the vocal tract of the singer. The features we develop in this document capture long-term information related to the intonation of the singer, which is relevant to the style and the technique of the singer. We propose a method to combine these two complementary descriptions of the singing voice to increase the recognition rate of singer identification. In addition we evaluate the robustness of each type of feature against a set of variations. We show the singing voice is a highly variable instrument. To obtain a representative model of a singer's voice it is thus necessary to build models using a large set of examples covering the full tessitura of a singer. In addition, we show that features extracted directly from the partials are more robust to the presence of an instrumental accompaniment than features derived from the amplitude spectrum.

Algunas aplicaciones de carretera actuales, tales como los servicios de información al viajero, llamadas de emergencia automáticas, control de flotas o telepeaje eletrónico, requieren una solución de calidad al problema del posicionamiento de un vehículo terrestre, que funcione en cualquier entorno y a un coste razonable. Esta tesis presenta una solución a este problema, fusionando para ello la información procedente principalmente de sensores de navegación por satélite y sensores inerciales. Para ello emplea un nuevo filtro de fusion multisensorial IMM-EKF. El comportamiento del sistema ha sido analizado en entornos reales y controlados, y comparado con otras soluciones propuestas. Finalmente, su aplicabilidad al problema planteado ha sido verificada. / Road applications such as traveller information, automatic emergency calls, freight management or electronic fee, collection require a onboard equipment (OBE) capable to offer a high available accurate position, even in unfriendly environments with low satellite visibility at low cost. Specifically in life critical applications, users demand from the OBEs accurate continuous positioning and information of the reliability of this position. This thesis presents a solution based on the fusion of Global Navigation Satellite Systems (GNSS) and inertial sensors (GNSS/INS), running an Extended Kalman Filter combined with an Interactive Multi-Model method (IMM-EKF). The solution developed in this work supplies continuous positioning in marketable conditions, and a meaningful trust level of the given solution. A set of tests performed in controlled and real scenarios proves the suitability of the proposed IMM-EKF implementation, as compared with low cost GNSS based solutions, dead reckoning systems and single model extended Kalman filter (SM-EKF) solutions.

Durant dècades s'ha observat i monitoritzat sistemàticament la Terra i el seu entorn des de l'espai o a partir de plataformes aerotransportades. Paral·lelament, s'ha tractat d'extreure el màxim d'informació qualitativa i quantitativa de les observacions realitzades. Les tècniques de fusió de dades donen un "ventall de procediments que ens permeten aprofitar les dades heterogènies obtingudes per diferents mitjans i instruments i integrar-les de manera que el resultat final sigui qualitativament superior". En aquesta tesi s'han desenvolupat noves tècniques que es poden aplicar a l'anàlisi de dades multiespectrals que provenen de sensors remots, adreçades a aplicacions oceanogràfiques. Bàsicament s'han treballat dos aspectes: les tècniques d'enregistrament o alineament d'imatges; i la interpolació de dades esparses i multiescalars, focalitzant els resultats als camps vectorials bidimensionals.En moltes aplicacions que utilitzen imatges derivades de satèl·lits és necessari mesclar o comparar imatges adquirides per diferents sensors, o bé comparar les dades d'un sòl sensor en diferents instants de temps, per exemple en: reconeixement, seguiment i classificació de patrons o en la monitorització mediambiental. Aquestes aplicacions necessiten una etapa prèvia d'enregistrament geomètric, que alinea els píxels d'una imatge, la imatge de treball, amb els píxels corresponents d'una altra imatge, la imatge de referència, de manera que estiguin referides a uns mateixos punts. En aquest treball es proposa una aproximació automàtica a l'enregistrament geomètric d'imatges amb els contorns de les imatges; a partir d'un mètode robust, vàlid per a imatges mutimodals, que a més poden estar afectades de distorsions, rotacions i de, fins i tot, oclusions severes. En síntesi, s'obté una correspondència punt a punt de la imatge de treball amb el mapa de referència, fent servir tècniques de processament multiresolució. El mètode fa servir les mesures de correlació creuada de les transformades wavelet de les seqüències que codifiquen els contorns de la línia de costa. Un cop s'estableix la correspondència punt a punt, es calculen els coeficients de la transformació global i finalment es poden aplicar a la imatge de treball per a enregistrar-la respecte la referència.A la tesi també es prova de resoldre la interpolació d'un camp vectorial espars mostrejat irregularment. Es proposa un algorisme que permet aproximar els valors intermitjos entre les mostres irregulars si es disposa de valors esparsos a escales de menys resolució. El procediment és òptim si tenim un model que caracteritzi l'esquema multiresolució de descomposició i reconstrucció del conjunt de dades. Es basa en la transformada wavelet discreta diàdica i en la seva inversa, realitzades a partir d'uns bancs de filtres d'anàlisi i síntesi. Encara que el problema està mal condicionat i té infinites solucions, la nostra aproximació, que primer treballarem amb senyals d'una dimensió, dóna una estratègia senzilla per a interpolar els valors d'un camp vectorial bidimensional, utilitzant tota la informació disponible a diferents resolucions. Aquest mètode de reconstrucció es pot utilitzar com a extensió de qualsevol interpolació inicial. També pot ser un mètode adequat si es disposa d'un conjunt de mesures esparses de diferents instruments que prenen dades d'una mateixa escena a diferents resolucions, sense cap restricció en les característiques de la distribució de mesures. Inicialment cal un model dels filtres d'anàlisi que generen les dades multiresolució i els filtres de síntesi corresponents, però aquest requeriment es pot relaxar parcialment, i és suficient tenir una aproximació raonable a la part passa baixes dels filtres. Els resultats de la tesi es podrien implementar fàcilment en el flux de processament d'una estació receptora de satèl·lits, i així es contribuiria a la millora d'aplicacions que utilitzessin tècniques de fusió de dades per a monitoritzar paràmetres mediambientals. / During the last decades a systematic survey of the Earth environment has been set up from many spatial and airborne platforms. At present, there is a continuous effort to extract and combine the maximum of quantitative information from these different data sets, often rather heterogeneous. Data fusion can be defined as "a set of means and tools for the alliance of data originating from different sources with the aims of a greater quality result". In this thesis we have developed new techniques and schemes that can be applied on multispectral data obtained from remote sensors, with particular interest in oceanographic applications. They are based on image and signal processing. We have worked mainly on two topics: image registration techniques or image alignment; and data interpolation of multiscale and sparse data sets, with focus on two dimensional vector fields. In many applications using satellite images, and specifically in those related to oceanographic studies, it is necessary to merge or compare multiple images of the same scene acquired from different captors or from one captor but at different times. Typical applications include pattern classification, recognition and tracking, multisensor data fusion and environmental monitoring. Image registration is the process of aligning the remotely sensed images to the same ground truth and transforming them into a known geographic projection (map coordinates). This step is crucial to correctly merge complementary information from multisensor data. The proposed approach to automatic image registration is a robust method, valid for multimodal images affected by distortions, rotations and, to a reasonably extend, with severe data occlusion. We derived a point to point matching of one image to a georeferenced map applying multiresolution signal processing techniques. The method is based on the contours of images: it uses a maximum cross correlation measure on the biorthogonal undecimated discrete wavelet transforms of the codified coastline contours sequences. Once this point to point correspondence is established, the coefficients of a global transform could be calculated and finally applied on the working image to register it to the georeferenced map. The second topic of this thesis focus on the interpolation of sparse irregularly-sampled vector fields when these sparse data belong to different resolutions. It is proposed a new algorithm to iteratively approximate the intermediate values between irregularly sampled data when a set of sparse values at coarser scales is known. The procedure is optimal if there is a characterized model for the multiresolution decomposition / reconstruction scheme of the dataset. The scheme is based on a fast dyadic wavelet transform and on its inversion using a filter bank analysis/synthesis implementation for the wavelet transform model. Although the problem is ill-posed, and there are infinite solutions, our approach, firstly worked for one dimension signals, gives an easy strategy to interpolate the values of a vector field using all the information available at different scales. This reconstruction method could be used as an extension on any initial interpolation. It can also be suitable in cases where there are sparse measures from different instruments that are sensing the same scene simultaneously at several resolutions, without any restriction to the characteristics of the data distribution. Initially a filter model for the generation of multiresolution data and their synthesis counterpart is the main requisite but; this assumption can be partially relaxed with the only requirement of a reasonable approximation to the low pass counterpart. The thesis results can be easily implemented on the process stream of any satellite receiving station and therefore constitute a first contribution to potential applications on data fusion of environmental monitoring.

Recent advances in VLSI technology, and wireless communication have enabled the development of tiny, low-cost sensor nodes that communicate over short distances. These sensor nodes, which consist of sensing, data processing, and wireless communication capabilities, suggest the idea of sensor networks based on collaborative effort of a large number of sensor nodes. Sensor networks hold the promise for numerous applications such as intrusion detection, weather monitoring, security and tactical surveillance, distributed computing, and disaster management. Several new protocols and algorithms have been proposed in the recent past in order to realize these applications. In this thesis, we consider the problem of routing in Wireless Sensor Networks (WSNs). Routing is a challenging problem in WSNs due to the inherent characteristics which distinguish these networks from the others. Several routing algorithms have been proposed for WSNs, each considering a specific network performance objective such as long network lifetime (ChangandTassiulas,2004), end-to-end delay guarantees (T.Heetal,2003), and data fusion (RazvanCristescuetal,2005) etc. In this thesis, we utilize the Potential-based Routing Paradigm to develop routing algorithms for different performance objectives of interest in WSNs. The basic idea behind the proposed approach is to assign a scalar called the potential to every sensor node in the network. Data is then forwarded to the neighbor with highest potential. Potentials cause the data to flow along certain paths. By defining potential fields appropriately, one can cause data to flow along preferred paths, so that the given performance objective is achieved. We have demonstrated the usefulness of this approach by considering three performance objectives, and defining potentials appropriately in each case. The performance objectives that we have considered are (i) maximizing the time to network partition, (ii) maximizing the packet delivery ratio, and (iii) Data fusion. In an operational sensor network, sensor nodes’ energy levels gradually deplete, leading eventually to network partition. A natural objective is to route packets in such a way that the time to network partition is maximized. We have developed a potential function for this objective. We analyzed simple network cases and used the insight to develop a potential function applicable to any network. Simulation results showed that considerable improvements in time to network partition can be obtained compared to popular approaches such as maximum lifetime routing, and shortest hop count routing. In the next step, we designed a potential function that leads to routes with high packet delivery ratios. We proposed a “channel-state aware” potential definition for a simple 2-relay network and performed a Markov-chain based analysis to obtain the packet delivery ratio. Considerable improvement was observed compared to a channel-state-oblivious policy. This motivated us to define a channel-state-dependent potential function for a general network. Simulation results showed that for a relatively slowly changing wireless network, our approach can provide up to 20% better performance than the commonly-used shortest-hop-count routing. Finally, we considered the problem of correlated data gathering in sensor networks. The routing approach followed in literature is to construct a spanning tree rooted at the sink. Every node in the tree aggregates its data with the data from its children in order to reduce the number of transmitted bits. Due to this fact, the total energy cost of the data collection task is a function of the underlying tree structure. Noting that potential based routing schemes also result in a tree structure, we present a potential definition that results in the minimum energy cost tree under some special conditions. Specifically, we consider a scenario in which sensor nodes’ measurements are quantized to K values. The task at the sink is to construct a histogram of measurements of all sensor nodes. Sensor nodes do not directly send their measurements to sink. Instead, they construct a temporary histogram using the data from its children and forward it to its parent node in the tree. We present a potential definition that results in the minimum energy cost tree under some conditions on sensor nodes’ measurements. We include both the transmission energy cost as well as the energy cost associated with the aggregation process.