Tesis sobre el tema "Bit (ordinateur)"
Crea una cita precisa en los estilos APA, MLA, Chicago, Harvard y otros
Consulte los 50 mejores tesis para su investigación sobre el tema "Bit (ordinateur)".
Junto a cada fuente en la lista de referencias hay un botón "Agregar a la bibliografía". Pulsa este botón, y generaremos automáticamente la referencia bibliográfica para la obra elegida en el estilo de cita que necesites: APA, MLA, Harvard, Vancouver, Chicago, etc.
También puede descargar el texto completo de la publicación académica en formato pdf y leer en línea su resumen siempre que esté disponible en los metadatos.
Explore tesis sobre una amplia variedad de disciplinas y organice su bibliografía correctamente.
Preda, Marius. "Système d'animation d'objets virtuels". Phd thesis, Paris 5, 2002. http://www.theses.fr/2002PA05S019.
Texto completoWithin the framework of the emerging networked multimedia information society, this dissertation brings forward methodological and technical contributions for representing, animating and transmitting 3D virtual objects. The existing methods are first analyzed and compared, and the performances of current multimedia standards are evaluated in terms of animation realism and transmission bitrate. .
Thiele, Stefan. "Read-out and coherent manipulation of an isolated nuclear spin using a single-molecule magnet spin-transistor". Phd thesis, Université de Grenoble, 2014. http://tel.archives-ouvertes.fr/tel-00984973.
Texto completoLi, Fangyan. "Simulation multi-moteurs multi-niveaux pour la validation des spécifications système et optimisation de la consommation". Thesis, Nice, 2016. http://www.theses.fr/2016NICE4008/document.
Texto completoThis work aims at system-level modelling a defined transceiver for Bluetooth Low energy (BLE) system using SystemC-AMS. The goal is to analyze the relationship between the transceiver performance and the accurate energy consumption. This requires the transceiver model contains system-level simulation speed and the low-level design block power consumption and other RF specifications. The Meet-in-the-Middle approach and the Baseband Equivalent method are chosen to achieve the two requirements above. A global simulation of a complete BLE system is achieved by integrating the transceiver model into a SystemC-TLM described BLE system model which contains the higher-than-PHY levels. The simulation is based on a two BLE devices communication system and is run with different BLE use cases. The transceiver Bit-Error-Rate and the energy estimation are obtained at the end of the simulation. First, we modelled and validated each block of a BT transceiver. In front of the prohibitive simulation time, the RF blocks are rewritten by using the BBE methodology, and then refined in order to take into account the non-linearities, which are going to impact the couple consumption, BER. Each circuit (each model) is separately verified, and then a first BLE system simulation (point-to-point between a transmitter and a receiver) has been executed. Finally, the BER is finally estimated. This platform fulfills our expectations, the simulation time is suitable and the results have been validated with the circuit measurement offered by Riviera Waves Company. Finally, two versions of the same transceiver architecture are modelled, simulated and compared
Melul, Franck. "Développement d'une nouvelle génération de point mémoire de type EEPROM pour les applications à forte densité d'intégration". Electronic Thesis or Diss., Aix-Marseille, 2022. http://www.theses.fr/2022AIXM0266.
Texto completoThe objective of this thesis was to develop a new generation of EEPROM memory for high reliability and high density applications. First, an innovative memory cell developed by STMicroelectronics - eSTM (Split-gate charge storage memory with buried vertical selection transistor) - was studied as a reference cell. In a second part, to improve the reliability of the eSTM cell and to allow a more aggressive miniaturization of the EEPROM cell, a new memory architecture has been proposed: the BitErasable cell. It showed an excellent reliability and allowed to bring elements of under-standing on the degradation mechanisms present in these memory devices with buried selection transistor. This new architecture also offers the possibility to individually erase cells in a memory array: bit by bit. Aware of the great interest of bit-by-bit erasing, a new erasing mechanism by hot hole injection has been proposed for the eSTM cell. It has shown performances and a level of reliability perfectly compatible with the industrial requirements of Flash-NOR applications
Petit, Sophie. "Étude des méthodes de prédiction de taux d'erreurs en orbite dans les mémoires : nouvelle approche empirique". Toulouse, ENSAE, 2006. http://www.theses.fr/2006ESAE0015.
Texto completoMarcireau, Alexandre. "Vision par ordinateur évènementielle couleur : cadriciel, prototype et applications". Thesis, Sorbonne université, 2019. http://www.theses.fr/2019SORUS248.
Texto completoNeuromorphic engineering is a bio-inspired approach to sensors and computers design. It aims to mimic biological systems down to the transistor level, to match their unparalleled robustness and power efficiency. In this context, event-based vision sensors have been developed. Unlike conventional cameras, they feature independent pixels which asynchronously generate an output upon detecting changes in their field of view, with high temporal precision. These properties are not leveraged by conventional computer vision algorithms, thus a new paradigm has been devised. It advocates short calculations performed on each event to mimic the brain, and shows promise both for computer vision and as a model of biological vision. This thesis explores event-based computer vision to improve our understanding of visual perception and identify potential applications. We approach the issue through color, a mostly unexplored aspect of event-based sensors. We introduce a framework supporting color events, as well as two experimental devices leveraging it: a three-chip event-based camera performing absolute color measurements, and a visual psychophysics setup to study the role of precise-timing in the brain. We explore the possibility to apply the color sensor to the genetic engineering Brainbow method, and present a new mathematical model for the latter
Terrissa, Saddek Labib. "Vision bio-fondée appliquée à l'analyse d'images". Le Havre, 2006. http://www.theses.fr/2006LEHA0001.
Texto completoThe environment recognition goes inevitably through the retina of alive beings. This component is a part of the central nervous system witch performs many processing, especially in motion detection and features extraction of a scene. These aspects are not reach the development level compared with static processing in artificial vision. For these reasons, we are developed a generic prototype of artificial retina inspired from biology witch we are titled Retina. This system is based on artificial neural network organised to simulate the radial variation of the acuity. The information is encoded through the implementation of analogue and impulse neurons. These retinal signal impulses are analysed in order to detect the variation of grey level in dynamic images. Moreover, we reconstitute these images using neural networks techniques and signal processing
Debaecker, Thibaud. "Geometric and bio-inspired analysis of non-linear image sensors". Paris 6, 2010. http://www.theses.fr/2010PA066717.
Texto completoLeclercq, Edouard. "Vision artificielle par modules bio-fondés spécifiques : forme et mouvement". Le Havre, 1999. http://www.theses.fr/1999LEHA0001.
Texto completoLee, Yu Chung Eugene. "Co-ordinated supply chain management and optimal control problems". online access from Digital Dissertation Consortium, 2007. http://libweb.cityu.edu.hk/cgi-bin/er/db/ddcdiss.pl?3299869.
Texto completoVandevuer, Stéphane. "Etude assistée par ordinateur de protéines et de leurs interactions avec des ligands :application à la D-alanine:Dalanine ligase et à la P-glycoprotéine". Doctoral thesis, Universite Libre de Bruxelles, 2006. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/210792.
Texto completoGiovannangeli, Christophe. "Navigation bio-mimétique autonome en environnements intérieurs et extérieurs : apprentissage sensori-moteur et planification dans un cadre interactif". Cergy-Pontoise, 2007. http://biblioweb.u-cergy.fr/theses/07CERG0346.pdf.
Texto completoThis thesis aims at highlighting how the interplay between robotics and the sciences of the living can promote both the understanding the cerebral circuitry and the achievment of very robust algorithms for robotics. The problem of the development of a neural achitecture, inspired by neurobiological models of the cerebral circuitry, for autonomous navigation of mobile robots in unknown indoor and outdoor environments, is adressed. A localization system based on a model of purely visual place-cells is presented. The place-cells enable to learn reactive sensory-motor behavior and planning behavior. The system is able to learn behavioral tasks autonomously or by interacting with a human teacher. An orientation system, merging a visual compass and path integration information, will also be proposed. The possible interactions between sensory-motor and planning strategies will be debated. Finally, issues on the autonomy of the learning in robot will be adressed: A metacontrol architecture which gives self-evaluation capabilities by analyzing the learning progress will be presented
Grard, Matthieu. "Generic instance segmentation for object-oriented bin-picking". Thesis, Lyon, 2019. http://www.theses.fr/2019LYSEC015.
Texto completoReferred to as robotic random bin-picking, a fast-expanding industrial task consists in robotizing the unloading of many object instances piled up in bulk, one at a time, for further processing such as kitting or part assembling. However, explicit object models are not always available in many bin-picking applications, especially in the food and automotive industries. Furthermore, object instances are often subject to intra-class variations, for example due to elastic deformations.Object pose estimation techniques, which require an explicit model and assume rigid transformations, are therefore not suitable in such contexts. The alternative approach, which consists in detecting grasps without an explicit notion of object, proves hardly efficient when the object geometry makes bulk instances prone to occlusion and entanglement. These approaches also typically rely on a multi-view scene reconstruction that may be unfeasible due to transparent and shiny textures, or that reduces critically the time frame for image processing in high-throughput robotic applications.In collaboration with Siléane, a French company in industrial robotics, we thus aim at developing a learning-based solution for localizing the most affordable instance of a pile from a single image, in open loop, without explicit object models. In the context of industrial bin-picking, our contribution is two-fold.First, we propose a novel fully convolutional network (FCN) for jointly delineating instances and inferring the spatial layout at their boundaries. Indeed, the state-of-the-art methods for such a task rely on two independent streams for boundaries and occlusions respectively, whereas occlusions often cause boundaries. Specifically, the mainstream approach, which consists in isolating instances in boxes before detecting boundaries and occlusions, fails in bin-picking scenarios as a rectangle region often includes several instances. By contrast, our box proposal-free architecture recovers fine instance boundaries, augmented with their occluding side, from a unified scene representation. As a result, the proposed network outperforms the two-stream baselines on synthetic data and public real-world datasets.Second, as FCNs require large training datasets that are not available in bin-picking applications, we propose a simulation-based pipeline for generating training images using physics and rendering engines. Specifically, piles of instances are simulated and rendered with their ground-truth annotations from sets of texture images and meshes to which multiple random deformations are applied. We show that the proposed synthetic data is plausible for real-world applications in the sense that it enables the learning of deep representations transferable to real data. Through extensive experiments on a real-world robotic setup, our synthetically trained network outperforms the industrial baseline while achieving real-time performances. The proposed approach thus establishes a new baseline for model-free object-oriented bin-picking
Emde, Claudia. "A polarized discrete ordinate scattering model for radiative transfer simulations in spherical atmospheres with thermal source". Berlin Logos-Verl, 2005. http://deposit.ddb.de/cgi-bin/dokserv?idn=975294377.
Texto completoEscobar-Zuniga, María-José. "Modèles bio-inspirés pour l'estimation et l'analyse de mouvement : reconnaissance d'actions et intégration du mouvement". Nice, 2009. http://www.theses.fr/2009NICE4050.
Texto completoThis thesis addresses the study of the motion perception in mammals and how bio-inspired systems can be applied to real applications. The first part of this thesis relates how the visual information is processed in the mammal's brains and how motion estimation is usually modeled. Based on this analysis of the state of the art, we propose a feedforward V1-MT core architecture. This feedforward V1-MT core architecture will be a basis to study two different kinds of applications. The first application is human action recognition, which is still a challenging problem in the computer vision community. We show how our bio-inspired method can be successfully applied to this real application. Interestingly, we show how several computational properties inspired from motion processing in mammals, allow us to reach high quality results, which will be compared to latest reference results. The second application of the bio-inspired architecture proposed in this thesis, is to consider the problem of motion integration for the solution of the aperture problem. We investigate the role of delayed V1 surround suppression, and how the 2D information extracted through this mechanism can be integrated to propose a solution for the aperture problem. Finally, we highlight a variety of important issues in the determination of motion estimation and additionally we present many potential avenues for future research efforts
Jadawi, Akram. "Expanded bio-thermoplastic foam obtained from starch : thermo-physical and mechanical characterizations". Rouen, 2014. http://www.theses.fr/2014ROUES019.
Texto completoBovo, Angela. "Apprentissage automatique pour l'assistance au suivi d'étudiants en ligne : approches classique et bio-inspirée". Thesis, Toulouse 1, 2014. http://www.theses.fr/2014TOU10035/document.
Texto completoThis Ph.D. took the shape of a partnership between the VORTEX team in the computer science research laboratory IRIT and the company Andil, which specializes in software for e-learning. This partnership was concluded around a CIFRE Ph.D. This plan is subsidized by the French state through the ANRT. The Ph.D. student, Angela Bovo, worked in Université Toulouse 1 Capitole. Another partnership was built with the training institute Juriscampus, which gave us access to data from real trainings for our experiments. Our main goal for this project was to improve the possibilities for monitoring students in an e-learning training to keep them from falling behind or giving up. We proposed ways to do such monitoring with classical machine learning methods, with the logs from students' activity as data. We also proposed, using the same data, indicators of students' behaviour. With Andil, we designed and produced a web application called GIGA, already marketed and sold, and well appreciated by training managers, which implements our proposals and served as a basis for first clustering experiments which seem to identify well students who are failing or about to give up. Another goal of this project was to study the capacities of the human brain inspired machine learning algorithm Hierarchical Temporal Memory (HTM), in its Cortical Learning Algorithm (CLA) version, because its base hypotheses are well adapted to our problem. We proposed ways to adapt HTM-CLA to classical machine learning functionalities (clustering, classification, regression, prediction), in order to compare its results to those of more classical algorithms; but also to use it as a basis for a behaviour generation engine, which could be used to create an intelligent tutoring system tasked with advising students in real time. However, our implementations did not get to the point of conclusive results
Boisard, Olivier. "Optimization and implementation of bio-inspired feature extraction frameworks for visual object recognition". Thesis, Dijon, 2016. http://www.theses.fr/2016DIJOS016/document.
Texto completoIndustry has growing needs for so-called “intelligent systems”, capable of not only ac-quire data, but also to analyse it and to make decisions accordingly. Such systems areparticularly useful for video-surveillance, in which case alarms must be raised in case ofan intrusion. For cost saving and power consumption reasons, it is better to perform thatprocess as close to the sensor as possible. To address that issue, a promising approach isto use bio-inspired frameworks, which consist in applying computational biology modelsto industrial applications. The work carried out during that thesis consisted in select-ing bio-inspired feature extraction frameworks, and to optimize them with the aim toimplement them on a dedicated hardware platform, for computer vision applications.First, we propose a generic algorithm, which may be used in several use case scenarios,having an acceptable complexity and a low memory print. Then, we proposed opti-mizations for a more global framework, based on precision degradation in computations,hence easing up its implementation on embedded systems. Results suggest that whilethe framework we developed may not be as accurate as the state of the art, it is moregeneric. Furthermore, the optimizations we proposed for the more complex frameworkare fully compatible with other optimizations from the literature, and provide encourag-ing perspective for future developments. Finally, both contributions have a scope thatgoes beyond the sole frameworks that we studied, and may be used in other, more widelyused frameworks as well
Alonzo, Flavien. "Méthodes numériques pour le Glioblastome Multiforme et pour la résolution de problèmes inverses autour des systèmes de réaction-diffusion". Electronic Thesis or Diss., Ecole centrale de Nantes, 2022. http://www.theses.fr/2022ECDN0059.
Texto completoGlioblastoma Multiforme is the most frequent and deadliest brain tumour. Mathematics stand as an innovative tool to enhance patient care in the context of personalized medicine. This PhD showcases two major contribution to this theme. A first contribution works on the modelling and simulating of a realistic spreading of the tumour cells in Glioblastoma Multiforme from a patient’s diagnosis. This work models tumour induced angiogenesis. A numerical scheme and algorithmare used to ensure positivity of solutions. Finally, simulations are compared to empirical knowledge from Medicine. A second contribution is on parameter estimation for reaction-diffusion models. The developed method solves inverse problems by solving two partial differential equation systems with a functional constraint, without using statistical tools. Numerical resolution of such problems is given and showcased on two examples of models with synthetic data. This method enables to calibrate parameters from a model using sparse data in time
Steyer, Jean-Philippe. "Sur une approche qualitative des systemes physiques. Aide en temps reel a la conduite des proceds fermentaires". Toulouse 3, 1991. http://www.theses.fr/1991TOU30251.
Texto completoBrégier, Romain. "Détection et estimation de pose d'instances d'objet rigide pour la manipulation robotisée". Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAM039.
Texto completoVisual object detection and estimation of their poses -- i.e. position and orientation for a rigid object -- is of utmost interest for automatic scene understanding.In this thesis, we address this topic through the bin-picking scenario, in which instances of a rigid object have to be automatically detected and localized in bulk, so as to be manipulated by a robot for various industrial tasks such as machine feeding, assembling, packing, etc.To this aim, we propose a novel method for object detection and pose estimation given an input depth image, based on the aggregation of local predictions through an Hough forest technique, that is suitable with industrial constraints of performance and ease of use.Overcoming limitations of existing approaches that assume objects not to have any proper symmetries, we develop a theoretical and practical framework enabling us to consider any physical rigid object, thanks to a novel definition of the notion of pose and an associated distance.This framework provides tools to deal with poses efficiently for operations such as pose averaging or neighborhood queries, and is based on rigorous mathematical developments.Evaluation benchmarks used in the literature are not very representative of our application scenario and suffer from some intrinsic limitations, therefore we formalize a methodology suited for scenes in which many object instances, partially occluded, in arbitrary poses may be considered. We apply this methodology on real and synthetic data, and demonstrate the soundness of our approach compared to the state of the art
Basu, Ananda Shankar. "Modélisation à base de composants de systèmes temps réel hétérogènes en BIP". Grenoble 1, 2008. http://www.theses.fr/2008GRE10249.
Texto completoWe present the BIP component framework for component-based construction of real-time systems from heterogeneous components. BIP stands for Behavior, Interaction and Priority, the three main elements for building components. It considers that systems can be obtained by composition of 3-layered components, where the lowest layer represents behavior as a set of transitions with triggers and actions; the intermediate level is the set of the interactions between the transitions of the behavior level, and the upper level is a set of dynamic priority rules. BIP has a rigorous formal semantics, and supports a system construction methodology based on a parameterized composition operator on components. The use of a layered composition operator allows incremental construction. The system construction process can be viewed as a sequence of transformations in a three-dimensional space of Behavior X Interaction X Priority, and provides a basis for the study of property preserving transformations between subclasses of systems such as untimed/timed, asynchronous/synchronous. We also provide a distributed semantics for systems in BIP, using a partial state model, and show the use of an auxiliary predicate called Oracle to preserve observational equivalence with respect to the centralized semantics. We present an implementation of BIP as a tool chain that allows modeling, static analysis and execution of BIP descriptions on a dedicated engine. Three different implementations for the engine have been provided, a centralized enumerative engine, a centralized symbolic engine and a distributed engine. We illustrate the use of the tool chain for two major application domains, Software Componentization, and Modeling mixed hw/sw systems
Rao, Balappa Shrisha. "Fine structure in cortical connectivity : effects on network dynamics and function Dynamics and orientation selectivity in a cortical model of rodent V1 with excess bidirectional connections Theory of orientation selectivity in random binary networks". Thesis, Sorbonne Paris Cité, 2018. https://wo.app.u-paris.fr/cgi-bin/WebObjects/TheseWeb.woa/wa/show?t=2401&f=17357.
Texto completoThe local cortical network connectivity significantly deviates from a random network, giving rise to fine structure at the neuron-to-neuron level. In this study, we have investigated the effects of these fine structures on network dynamics and function. We have investigated two types of fine structure, namely, excess bidirectionality and feature specific connectivity. The study of the effects of excess bidirectionality was conducted in a conductance-based model of layer 2/3 in rodent V1. Through large scale numerical simulations, we showed that excess bidirectional connections in the inhibitory population leads to slower dynamics. Remarkably, we found that bidirectional connections between inhibitory cells are more efficacious in slowing down the dynamics than those between the excitatory cells. Additionally, bidirectional connections between inhibitory cells increases the trial-to-trial variability, while between the excitatory and inhibitory populations it reduces the variability leading to improved coding efficiency. Our results suggest that the strong reciprocal connections between excitatory and PV+ cells that have been experimentally reported can improve coding efficiency by reducing the signal-to-noise ratio. The second part of this work involved an analytical study of a model of layer 2/3 rodent V1 with binary neurons. In our study, we assumed that neurons in layer 4 were selective to stimuli orientation. Our results account for the changes in tuning properties observed during the critical period in mouse V1. Prior to the critical period, the connectivity between pyramidal neurons in the mouse V1 is non-specific. Following previous studies of spiking networks, we analytically demonstrated that with such connectivity, layer 2/3 neurons in our model develop orientation selectivity. A small fraction of strong feature specific connections between pyramidal cells have been reported in the mouse V1 after the critical period. We showed that, in spite of their small number, such connections can substantially impact the tuning of layer 2/3 cells to orientation: excitatory neurons become more selective and through non-specific global changes in their synaptic strengths, the inhibitory cells become more broadly tuned
Loquet, Jean-Gabriel. "Étude numérique et expérimentale des dommages permanents induits par une particule lourde dans les composants électroniques". École nationale supérieure de l'aéronautique et de l'espace (Toulouse ; 1972-2007), 2001. http://www.theses.fr/2001ESAE0015.
Texto completoGouache, Thibault. "Forage alternatif dual de régolithe extra-terrestre : évolution d'une solution bio-inspirée". Toulouse, ISAE, 2011. http://www.theses.fr/2011ESAE0011.
Texto completoStaub, Matthias. "Approche multi-échelle du comportement bio-mécanique d'un déchet non dangereux". Grenoble, 2010. http://www.theses.fr/2010GRENU025.
Texto completoThis research addresses the biomechanical evolution of municipal solid waste subject to different pretreatment and operational conditions at different scales. After an introduction to the major stakes and figures related to waste management, waste landfilling and its evolution (Chapter I), the characterization of the waste medium is addressed (Chapter II). This triphasic unsaturated medium requires a dedicated scientific approach as well as specific investigation experiments. Several past investigations have demonstrated the need for specific coupled studies of waste. To do so, it seems also essential to adapt measurement methods (moisture, density. . . ) to the waste medium (Chapter III). Hence, measurement techniques have been studied in detail and validated for their use in this medium at scales ranging from the laboratory to the site. An experimental platform consisting of four pilot cells at a semi-industrial scale, designed with Veolia Environnement Recherche & Innovation, has been used at LTHE to investigate waste biomechanics at a sufficient scale and under site-near conditions (compression, temperature, moisture. . . ) (Chapter IV). The results enable to characterize biodegradation in terms of daily monitoring as well as final budgets, but also to identify the major drivers of biodegradation depending on the pre-treatment and operational conditions. Finally, these results and other laboratory- and site-scale results have been used to demonstrate the biomechanical coupling and to propose a model for waste settlement (Chapter V)
Shahsavari, Mahyar. "Unconventional computing using memristive nanodevices : from digital computing to brain-like neuromorphic accelerator". Thesis, Lille 1, 2016. http://www.theses.fr/2016LIL10203/document.
Texto completoBy 2020, there will be 50 to 100 billion devices connected to the Internet. Two domains of hot research to address these high demands of data processing are the Internet of Things (IoT) and Big Data. The demands of these new applications are increasing faster than the development of new hardware particularly because of the slowdown of Moore's law. The main reason of the ineffectiveness of the processing speed is the memory wall or Von Neumann bottleneck which is coming from speed differences between the processor and the memory. Therefore, a new fast and power-efficient hardware architecture is needed to respond to those huge demands of data processing. In this thesis, we introduce novel high performance architectures for next generation computing using emerging nanotechnologies such as memristors. We have studied unconventional computing methods both in the digital and the analog domains. However, the main focus and contribution is in Spiking Neural Network (SNN) or neuromorphic analog computing. In the first part of this dissertation, we review the memristive devices proposed in the literature and study their applicability in a hardware crossbar digital architecture. At the end of part~I, we review the Neuromorphic and SNN architecture. The second part of the thesis contains the main contribution which is the development of a Neural Network Scalable Spiking Simulator (N2S3) suitable for the hardware implementation of neuromorphic computation, the introduction of a novel synapse box which aims at better learning in SNN platforms, a parameter exploration to improve performance of memristor-based SNN, and finally a study of the application of deep learning in SNN
Mercier, Michael. "Contribution to High Performance Computing and Big Data Infrastructure Convergence". Thesis, Université Grenoble Alpes (ComUE), 2019. http://www.theses.fr/2019GREAM031/document.
Texto completoThe amount of data produced, either in the scientific community and the commercial world, is constantly growing. The field of Big Data has emerged to handle a large amount of data on distributed computing infrastructures. High-Performance Computer (HPC) infrastructures are made for intensive parallel computations. The HPC community is also facing more and more data because of new high definition sensors and large physics apparatus. The convergence of the two fields is currently happening. In fact, the HPC community is already using Big Data tools, but they are not integrated correctly, especially at the level of the file system and the Resources and Job Management System (RJMS).In order to understand how we can leverage HPC clusters for Big Data usage, and what are the challenges for the HPC infrastructures, we have studied multiple aspects of the convergence: we have made a survey on the software provisioning methods, with a focus on data-intensive applications. We also propose a new RJMS collaboration technique called BeBiDa which is based on 50 lines of code whereas similar solutions use at least 1000x more. We evaluate this mechanismon real conditions and in a simulation with our simulator Batsim
Vernay, Rémi. "Etudes d'objets combinatoires : applications à la bio-informatique". Phd thesis, Université de Bourgogne, 2011. http://tel.archives-ouvertes.fr/tel-00668134.
Texto completoHu, Dawei. "Contribution à l’étude du comportement mécanique et hydromécanique d’une roche endommagée". Thesis, Lille 1, 2009. http://www.theses.fr/2009LIL10150/document.
Texto completoThis thesis presents the laboratory and numerical investigations on anisotropic mechanical, poromechanical and fluid transport behavior of sandstone. After the X-ray and SEM tests, the samples were stressed under triaxial loading to study the initial and stress-induced mechanical behavior. The evolution of anisotropic Biot’s coefficients and axial permeability are also investigated and related to mechanical behavior. A plastic damage model is presented based on discrete approach. The proposed model is applied to simulate the mechanical behavior of dry sandstone under different loading conditions, the evolution and distribution of damage variable and plastic hardening variable are also discussed. Furthermore, the proposed model is applied to simulate the previous experimental results (including mechanical behavior, Biot's coefficients and permeability) of saturated sandstones. The comparisons between the numerical result and test data show a good performance of the present model. Finally, an anisotropic poroplastic damage model is developed to describe the poromechanical behavior of saturated sandstone. The comparisons between numerical simulation and experimental data is given for the triaxial compression tests under drained and undrained condition, the tests of strain response to pore pressure increment is also simulated, a good accordance is obtained
Hemmati, Sahar. "Etude de l'Interaction Sol-Végétation-Atmosphère avec une approche couplée Thermo-Hydro-Mécanique". Phd thesis, Ecole des Ponts ParisTech, 2009. http://pastel.archives-ouvertes.fr/pastel-00005609.
Texto completoZaourar, Lilia Koutchoukali. "Recherche opérationnelle et optimisation pour la conception testable de circuits intégrés complexes". Grenoble, 2010. http://www.theses.fr/2010GRENM055.
Texto completoThis thesis is a research contribution interfacing operations research and microelectronics. It considers the use of combinatorial optimization techniques for DFT (Design For Test) of Integrated Circuits (IC). With the growing complexity of current IC both quality and cost during manufacturing testing have become important parameters in the semiconductor industry. To ensure proper functioning of the IC, the testing step is more than ever a crucial and difficult step in the overall IC manufacturing process. To answer market requirements, chip testing should be fast and effective in uncovering defects. For this, it becomes essential to apprehend the test phase from the design steps of IC. In this context, DFT techniques and methodologies aim at improving the testability of IC. In previous research works, several problems of optimization and decision making were derived from the micro- electronics domain. Most of previous research contributions dealt with problems of combinatorial optimization for placement and routing during IC design. In this thesis, a higher design level is considered where the DFT problem is analyzed at the Register Transfer Level (RTL) before the logic synthesis process starts. This thesis is structured into three parts. In the first part, preliminaries and basic concepts of operations research, IC design and manufacturing are introduced. Next, both our approach and the solution tools which are used in the rest of this work are presented. In the second part, the problem of optimizing the insertion of scan chains is considered. Currently, " internal scan" is a widely adopted DFT technique for sequential digital designs where the design flip-flops are connected into a daisy chain manner with a full controllability and observability from primary inputs and outputs. In this part of the research work, different algorithms are developed to provide an automated and optimal solution during the generation of an RTL scan architecture where several parameters are considered: area, test time and power consumption in full compliance with functional performance. This problem has been modelled as the search for short chains in a weighted graph. The solution methods used are based on finding minimal length Hamiltonian chains. This work was accomplished in collaboration with DeFacTo Technologies, an EDA start-up close to Grenoble. The third part deals with the problem of sharing BIST (Built In Self Test) blocks for testing memories. The problem can be formulated as follows: given the memories with various types and sizes, and sharing rules for series and parallel wrappers, we have to identify solutions to the problem by associating a wrapper with each memory. The solution should minimize the surface, the power consumption and test time of IC. To solve this problem, we designed a prototype called Memory BIST Optimizer (MBO). It consists of two steps of resolution and a validation phase. The first step creates groups of compatibility in accordance with the rules of abstraction and sharing that depend on technologies. The second phase uses genetic algorithms for multi-objective optimization in order to obtain a set of non dominated solutions. Finally, the validation verifies that the solution provided is valid. In addition, it displays all solutions through a graphical or textual interface. This allows the user to choose the solution that fits best. The tool MBO is currently integrated into an industrial flow within ST-microelectronics
Michaud, Dorian. "Indexation bio-inspirée pour la recherche d'images par similarité". Thesis, Poitiers, 2018. http://www.theses.fr/2018POIT2288/document.
Texto completoImage Retrieval is still a very active field of image processing as the number of available image datasets continuously increases.One of the principal objectives of Content-Based Image Retrieval (CBIR) is to return the most similar images to a given query with respect to their visual content.Our work fits in a very specific application context: indexing small expert image datasets, with no prior knowledge on the images. Because of the image complexity, one of our contributions is the choice of effective descriptors from literature placed in direct competition.Two strategies are used to combine features: a psycho-visual one and a statistical one.In this context, we propose an unsupervised and adaptive framework based on the well-known bags of visual words and phrases models that select relevant visual descriptors for each keypoint to construct a more discriminative image representation.Experiments show the interest of using this this type of methodologies during a time when convolutional neural networks are ubiquitous.We also propose a study about semi interactive retrieval to improve the accuracy of CBIR systems by using the knowledge of the expert users
Alric, Matthieu. "Conception et modélisation modulaire d'un robot bio-inspiré extensible pour l'accès aux tumeurs dans le cerveau". Phd thesis, Université Blaise Pascal - Clermont-Ferrand II, 2009. http://tel.archives-ouvertes.fr/tel-00724718.
Texto completoTremblay-Boucher, Vincent. "FLCAA : système de codage parcimonieux et d'analyse perceptuelle des signaux sonores en temps réel". Mémoire, Université de Sherbrooke, 2013. http://hdl.handle.net/11143/6215.
Texto completoKabalan, Hassan. "Minimizing Time Delay between Mobiles Exchanging Data in an Industrial Wireless Network Cell". Versailles-St Quentin en Yvelines, 2013. http://www.theses.fr/2013VERS0026.
Texto completoThe purpose of this study is to discuss the exchanges between mobiles moving in an industrial environment. To reach this study, a simulation approach has been chosen in order reach this study. This novel method aimed to minimize the exchange time between mobiles within an 802. 11 cell. The first step of our study represents the main points arising from this choice. Initially, we described the main existing models of spread and constraints of radio propagation. The selection of relevant model propagation for the selected industrial site is based on several measurements in domain indoor. The implementation of the ITU Pr1238 model was performed by adjusting the parameters of the various models; we decided to select the model recommended by the ITU under reference Pr1238 as our model that is used for the industrial by the NS2 Simulator. The second step was to come up with a modified method of the DCF in order to minimize the necessary time so that all mobiles within the same cell of an 802. 11b network can exchange the shared information during their co-operation. The optimization of the exchanges time between mobiles was carried out by modifying the binary exponential aspect of the Back-off algorithm as a first phase to reduce the access time to the radio medium. Whereas in the second phase this modified and enhanced BEB method was supported by a Neural Network Function to give us precise output parameters. Those outputs will be learned by the Neural Network Function and will be used in the NS2 simulation to get the new results of the time delay to compare them with the standard BEB method results
Koch, Mathilde. "Computational modeling to design and analyze synthetic metabolic circuits". Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLS467/document.
Texto completoThe aims of this thesis are two-fold, and centered on synthetic metabolic circuits, which perform sensing and computation using enzymes.The first part consisted in developing reinforcement and active learning tools to improve the design of metabolic circuits and optimize biosensing and bioproduction. In order to do this, a novel algorithm (RetroPath3.0) based on similarity-guided Monte Carlo Tree Search to improve the exploration of the search space is presented. This algorithm, combined with data-derived reaction rules and varying levels of enzyme promiscuity, allows to focus exploration on the most promising compounds and pathways for bio-retrosynthesis. As retrosynthesis-based pathways can be implemented in whole cell or cell-free systems, an active learning method to efficiently explore the combinatorial space of components for rational media optimization was also developed, to design the best media maximizing cell-free productivity.The second part consisted in developing analysis tools, to generate knowledge from biological data and model biosensor response. First, the effect of plasmid copy number on sensitivity of a transcription-factor based biosensor was modeled. Then, using cell-free systems allowing for broader control over the experimental factors such as DNA concentration, resource usage was modeled to ensure our current knowledge of underlying phenomenons is sufficient to account for circuit behavior, using either empirical models or mechanistic models. Coupled with metabolic circuit design, those models allowed us to develop a new biocomputation approach, called metabolic perceptrons.Overall, this thesis presents tools to design and analyse synthetic metabolic circuits, which are a novel way to perform computation in synthetic biology
Hedayat, Sara. "Conception et fabrication de neurones artificiels pour le traitement bioinspiré de l'information". Thesis, Lille 1, 2018. http://www.theses.fr/2018LIL1I039/document.
Texto completoCurrent computing technology has now reached its limits and it becomes thus urgent to propose new paradigms for information processing capable of reducing the energy consumption while improving the computing performances. Moreover, the human brain, is a fascinating and powerful organ with remarkable performances in areas as varied as learning, creativity, fault tolerance. Furthermore, with its total 300 billion cells, is able to perform complex cognitive tasks by consuming only around 20W. In this context, we investigated a new paradigm called neuromorphic or bio-inspired information processing.More precisely, the purpose of this thesis was to design and fabricate an ultra-low power artificial neuron using recent advances in neuroscience and nanotechnology. First, we investigated the functionalities of living neurons, their neuronal membrane and explored different membrane models known as Hodgkin Huxley, Wei and Morris Lecar models. Second, based on the Morris Lecar model, we designed analog spiking artificial neurons with different time constants and these neurons were fabricated using 65nm CMOS technology. Then we characterized these artificial neurons and obtained state of the art performances in terms of area, dissipated power and energy efficiency. Finally we investigated the noise within these artificial neurons, compared it with the biological sources of noise in a living neuron and experimentally demonstrated the stochastic resonance phenomenon. These artificial neurons can be extremely useful for a large variety of applications, ranging from data analysis (image and video processing) to medical aspect (neuronal implants)
Noé, Laurent. "Recherche de similarités dans les séquences d'ADN : modèles et algorithmes pour la conception de graines efficaces". Phd thesis, Université Henri Poincaré - Nancy I, 2005. http://tel.archives-ouvertes.fr/tel-00011482.
Texto completoElles se basent sur un principe de filtrage du texte qui permet de localiser les régions potentiellement similaires.
Dans cette thèse, nous proposons de nouvelles définitions de filtres pour la recherche de similarités sur les séquences génomiques et des algorithmes associés pour mesurer leurs caractéristiques.
Plus précisément, nous avons étudié le modèle des graines espacées, et proposé un algorithme d'évaluation de l'efficacité des graines sur des similarités d'une classe particulière (similarités dites homogènes). Nous avons également développé un algorithme général pour la mesure de l'efficacité des graines, ainsi qu'un nouveau modèle de graine appelé graine sous-ensemble, extension du modèle des graines espacées. Enfin nous donnons, dans le cadre du filtrage sans perte, une extension à l'aide de graines multiples, que nous analysons et appliquons au problème de la conception d'oligonucléotides.
Nous avons réalisé et donnons accès à des outils pour la conception des filtres, ainsi que pour la recherche de similarités.
Garnier, Simon. "Décisions collectives dans des systèmes d'intelligence en essaim". Toulouse 3, 2008. http://thesesups.ups-tlse.fr/386/.
Texto completoThrough an approach mixing ethology and robotics, experiments and simulations, we were interested in two self-organizing mechanisms used by two social insect species to collectively make decisions. Based on behavioural mimicry, these mechanisms most of the time lead individuals belonging to a group to reach a consensus. We studied in particular the influence of the physical structure of the environment on the final choice of the group. Our results shows that an asymmetry in the structure of the environment can be amplified by the underlying self-organizing process and therefore leads the group to select a particular alternative in most of the cases. As a consequence, taking into account the interaction between the behaviour of the group and the structure of the environment is of crucial importance during the study of animal societies and also during the design of artificial self-organized systems
Lefebvre, Sylvain. "Services de répartition de charge pour le Cloud : application au traitement de données multimédia". Electronic Thesis or Diss., Paris, CNAM, 2013. http://www.theses.fr/2013CNAM0910.
Texto completoThe research work carried out in this thesis consists in the development of new load balancing algorithms aimed at big data computing. The first algorithm, called « WACA » (Workload and Cache Aware Algorithm), enhances response times by locating data efficiently through content summaries. The second algorithm, called CAWA (Cost AWare Algorithm) takes advantage of the cost information available on Cloud Computing platforms by studying the workload history.Evaluation of these algorithms required the development of a cloud infrastructure simulator named Simizer, to enable testing of these policies prior to their deployment. This deployment can be transparently done thanks to the Cloudizer web service distribution and monitoring system, also developed during this thesis. These works are included in the Multimedia for Machine to Machine (MCUBE) project, where the Cloudizer Framework is deployed
Marion, Damien. "Sharing big display : développement des technologies et métaphores d'interactions nouvelles pour le partage collaboratif d'affichage en groupe ouvert". Thesis, Bordeaux, 2017. http://www.theses.fr/2017BORD0924/document.
Texto completoScreens have invaded our daily life. Amongst them, large displays are becoming increasingly present in public places. Sometimes, a few interaction are proposed to the users, but most of the time they are simply used as static displays. So, how to interact with these large displays and especially how to allow multi-user interactions? Of course, we have to define rules for collaboration: what we have to do if several people consult the same information at the same time? How to allow a newcomer to become aware of what there is to see when users have already "transformed" the display by their current use?First, our work consisted in testing instrumented multi-user interactions, based on a head tracking system, during a collaborative information seeking task, on a large display. In this part, we have highlighted the importance of the concept of "floating vision". Then, our research focused on the development of a head tracking system allowing intuitive interactions, without needing neither a special equipment nor individual calibration. Our system supports several users (up to six) simultaneously interacting with individualized information on a large display. Finally, we present the study of performance gain within a context of multi-user competitive consultation of information. We compare the benefit of an adaptive display (information move in front of users who are focusing on them), with a standard display. This study is based on user experience (UX) analysis. Thus, we were able to identify the first recommendations about interaction metaphors allowing intuitive interactions with a large display in an open group context.This research was devoted to highlighting the feasibility, interest and proposition of initial orientations concerning the design of multi-user interactions with large public displays based on a head tracking technique
Macher, Hélène. "Du nuage de points à la maquette numérique de bâtiment : reconstruction 3D semi-automatique de bâtiments existants". Thesis, Strasbourg, 2017. http://www.theses.fr/2017STRAD006/document.
Texto completoThe creation of an as-built BIM requires the acquisition of the as-is conditions of existing buildings. Terrestrial laser scanning (TLS) is widely used to achieve this goal. Indeed, laser scanners permit to collect information about object geometry in form of point clouds. They provide a large amount of accurate data in a very fast way and with a high level of details. Unfortunately, the scan-to-BIM process remains currently largely a manual process because of the huge amount of data and because of processes, which are difficult to automate. It is time consuming and error-prone. A key challenge today is thus to automate the process leading to 3D reconstruction of existing buildings from point clouds. The aim of this thesis is to develop a processing chain to extract the maximum amount of information from a building point cloud in order to integrate the result in a BIM software
El-khoury, Elie. "Unsupervised video indexing on audiovisuel characterization of persons". Toulouse 3, 2010. http://thesesups.ups-tlse.fr/1025/1/El-Khoury_Elie.pdf.
Texto completoThis thesis consists to propose a method for an unsupervised characterization of persons within audiovisual documents, by exploring the data related for their physical appearance and their voice. From a general manner, the automatic recognition methods, either in video or audio, need a huge amount of a priori knowledge about their content. In this work, the goal is to study the two modes in a correlated way and to explore their properties in a collaborative and robust way, in order to produce a reliable result as independent as possible from any a priori knowledge. More particularly, we have studied the characteristics of the audio stream and we have proposed many methods for speaker segmentation and clustering and that we have evaluated in a french competition. Then, we have carried a deep study on visual descriptors (face, clothing) that helped us to propose novel approches for detecting, tracking, and clustering of people within the document. Finally, the work was focused on the audiovisual fusion by proposing a method based on computing the cooccurrence matrix that allowed us to establish an association between audio and video indexes, and to correct them. That will enable us to produce a dynamic audiovisual model for each speaker
Abou, Rjeily Yves. "Management and sustainability of urban drainage systems within smart cities". Thesis, Lille 1, 2016. http://www.theses.fr/2016LIL10085/document.
Texto completoThis work presents the Real Time Control (RTC) of Urban Drainage Systems (UDS) within smart cities. RTC requires to understand the UDS operation and to perform simulations on measured, forecasted and synthetic events. Therefore, a Real Time Monitoring system (RTM) was implemented on the experimental site, and combined to a simulation model. A model auto-calibration process and hydraulic boundary conditions forecast system were developed, in order to simulate the hydrologic-hydraulic response. Aiming to protect the citizens and mitigate flooding consequences, the RTC was composed of a flooding forecast system followed by a dynamic management strategy. The proposed concept and methodologies were applied and evaluated on the Lille 1 University Campus, within the SunRise project. RTM was found very helpful in understanding the system operation and calibrating the simulation model. Genetic Algorithm followed by Pattern Search formed an effective auto-calibration procedure for the simulation model. NARX Neural Network was developed and validated for forecasting hydraulic boundary conditions. Once understanding the UDS operations, the RTC was developed. NARX Neural Network was found capable to forecast flooding events. A dynamic management for increasing a tank retention capacity, was studied based on calculating a Valve State Schedule, and results were satisfying by using Genetic Algorithm and a modified form of Artificial Bee Colony, as optimization methods. A qualitative management was also proposed and tested for verifying its potential in reducing flooding volumes
Yu, Mulin. "Reconstruction et correction de modèles urbains à l'aide de structures de données cinétiques". Thesis, Université Côte d'Azur, 2022. http://www.theses.fr/2022COAZ4077.
Texto completoCompact and accurate digital 3D models of buildings are commonly used by practitioners for the visualization of existing or imaginary environments, the physical simulations or the fabrication of urban objects. Generating such ready-to-use models is however a difficult problem. When created by designers, 3D models usually contain geometric errors whose automatic correction is a scientific challenge. When created from data measurements, typically laser scans or multiview images, the accuracy and complexity of the models produced by existing reconstruction algorithms often do not reach the requirements of the practitioners. In this thesis, I address this problem by proposing two algorithms: one for repairing the geometric errors contained in urban-specific formats of 3D models, and one for reconstructing compact and accurate models from input point clouds generated from laser scanning or multiview stereo imagery. The key component of these algorithms relies upon a space-partitioning data structure able to decompose the space into polyhedral cells in a natural and efficient manner. This data structure is used to both correct geometric errors by reassembling the facets of defect-laden 3D models, and reconstruct concise 3D models from point clouds with a quality that approaches those generated by Computer-Aided-Design interactive tools.My first contribution is an algorithm to repair different types of urban models. Prior work, which traditionally relies on local analysis and heuristic-based geometric operations on mesh data structures, is typically tailored-made for specific 3D formats and urban objects. We propose a more general method to process different types of urban models without tedious parameter tuning. The key idea lies on the construction of a kinetic data structure that decomposes the 3D space into polyhedra by extending the facets of the imperfect input model. Such a data structure allows us to re-build all the relations between the facets in an efficient and robust manner. Once built, the cells of the polyhedral partition are regrouped by semantic classes to reconstruct the corrected output model. I demonstrate the robustness and efficiency of the algorithm on a variety of real-world defect-laden models and show its competitiveness with respect to traditional mesh repairing techniques from both Building Information Modeling (BIM) and Geographic Information Systems (GIS) data.My second contribution is a reconstruction algorithm inspired by the Kinetic Shape Reconstruction method, that improves the later in different ways. In particular, I propose a data fitting technique for detecting planar primitives from unorganized 3D point clouds. Departing from an initial configuration, the technique refines both the continuous plane parameters and the discrete assignment of input points to them by seeking high fidelity, high simplicity and high completeness. The solution is found by an exploration mechanism guided by a multi-objective energy function. The transitions within the large solution space are handled by five geometric operators that create, remove and modify primitives. I demonstrate its potential, not on buildings only, but on a variety of scenes, from organic shapes to man-made objects
El, Asmi Emira. "Un environnement méthodologique et logiciel pour l'interopérabilité de la maquette numérique du bâtiment et de la simulation énergétique : application à la réglementation thermique RT2012". Thesis, Paris 8, 2016. http://www.theses.fr/2016PA080051.
Texto completoEnergy performance analysis of new buildings strongly relies on simulation tools. They optimize the design by facilitating building performance analysis during the design phase. However, there are several limitations, among which one of the main is the lack of reliable connections between Computer Aided Design (CAD) and simulation tools. This is particularly critical at a time when the concept of Building Information Model (BIM) is widely applied and the CAD models are becoming richer.This thesis proposes to take advantage of the potential of BIM in order to contribute to emerging approaches for a reliable and seamless interconnection between building information models and simulation environment input files. This would result into significant benefits, both from the cost-effectiveness (analysis is performed more effectively) and the reliability (simulation models strictly comply with design models) points of view. However, to implement such approaches, several barriers have to be considered. The first is the need for a common, standardized BIM language. The second is the definition of adequate and standardized extension mechanisms, to customize BIM for simulation-specific purposes. The third is the provision of software tools, for effective and reliable BIM to simulation models transformation.The first barrier is likely to be overcome through the advent of the buildingSMART Industry Foundation Classes (IFC), which is widely recognized as a standard BIM exchange format in the construction industry. However, with respect to the second barrier, the IFC remains a wide-purpose modelling language and as such, fails to address many domain-specific issues, like e.g. energy simulation. This limitation has been acknowledged by buildingSMART, who have delivered novel methodological tools to allow for domain-specific customization of IFC. The main one is called the Information Delivery Manual (IDM). The aim of our work is the implementation and evaluation the BuildingSMART framework IDM/MVD to interconnect BIM (IFC) and energy simulation. This study led us to propose a pivotal model (BSM for « Building Simulation Model») based on an analogy with model driven engineering approaches. We also have developed a software prototype, which includes a set of mapping rules that automates the translation process between the IFC model and the simulation tool. In the scope of this study, we targeted the COMETH simulation tool (a simulation engine based on the French thermal regulation RT2012) but our results can easily be extended to additional tools
Lefebvre, Sylvain. "Services de répartition de charge pour le Cloud : application au traitement de données multimédia". Phd thesis, Conservatoire national des arts et metiers - CNAM, 2013. http://tel.archives-ouvertes.fr/tel-01062823.
Texto completoMagnin, Loïck C. A. "Two-player interaction in quantum computing: cryptographic primitives and query complexity". Doctoral thesis, Universite Libre de Bruxelles, 2011. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/209783.
Texto completoPremièrement, elle étudie deux primitives cryptographiques quantiques, des briques de base pour construire des protocoles cryptographiques complexes entre deux joueurs, comme par exemple un protocole d'identification.
La première primitive est la "mise en gage quantique". Cette primitive ne peut pas être réalisée de manière inconditionnellement sûre, mais il est possible d'avoir une sécurité lorsque les deux parties sont soumises à certaines contraintes additionnelles. Nous étudions cette primitive dans le cas où les deux joueurs sont limités à l'utilisation d'états et d'opérations gaussiennes, un sous-ensemble de la physique quantique central en optique, donc parfaitement adapté pour la communication via fibres optiques. Nous montrons que cette restriction ne permet malheureusement pas la réalisation de la mise en gage sûre. Pour parvenir à ce résultat, nous introduisons la notion de purification intrinsèque, qui permet de contourner l'utilisation du théorème de Uhlman, en particulier dans le cas gaussien.
Nous examinons ensuite une primitive cryptographique plus faible, le "tirage faible à pile ou face", dans le modèle standard du calcul quantique. Carlos Mochon a donné une preuve d'existence d'un tel protocole avec un biais arbitrairement petit. Nous donnons une interprétation claire de sa preuve, ce qui nous permet de la simplifier et de la raccourcir grandement.
La seconde partie de cette thèse concerne l'étude de méthodes pour prouver des bornes inférieures dans le modèle de la complexité en requête. Il s'agit d'un modèle de complexité central en calcul quantique dans lequel de nombreux résultats majeurs ont été obtenus. Dans ce modèle, un algorithme ne peut accéder à l'entrée uniquement qu'en effectuant des requêtes sur chacune des variables de l'entrée. Nous considérons une extension de ce modèle dans lequel un algorithme ne calcule pas une fonction, mais doit générer un état quantique.
Cette généralisation nous permet de comparer les différentes méthodes pour prouver des bornes inférieures dans ce modèle. Nous montrons d'abord que la méthode par adversaire ``multiplicative" est plus forte que la méthode ``additive". Nous montrons ensuite une réduction de la méthode polynomiale à la méthode multiplicative, ce qui permet de conclure à la supériorité de la méthode par adversaire multiplicative sur toutes les autres méthodes.
Les méthodes par adversaires sont en revanche souvent difficiles à utiliser car elles nécessitent le calcul de normes de matrices de très grandes tailles. Nous montrons comment l'étude des symétries d'un problème simplifie grandement ces calculs.
Enfin, nous appliquons ces formules pour prouver la borne inférieure optimale du problème Index-Erasure, un problème de génération d'état quantique lié au célèbre problème Isomorphisme-de-Graphes.
Doctorat en Sciences de l'ingénieur
info:eu-repo/semantics/nonPublished
Clément, Michaël. "Modélisation et apprentissage de relations spatiales pour la reconnaissance et l’interprétation d’images". Electronic Thesis or Diss., Sorbonne Paris Cité, 2017. https://wo.app.u-paris.fr/cgi-bin/WebObjects/TheseWeb.woa/wa/show?t=1927&f=11772.
Texto completoIn recent years, the amount of visual data produced by various types of sensors has been continuously increasing. The automatic interpretation and indexation of such data constitute an important challenge in the fields of pattern recognition and computer vision. In this context, the relative position of the different objects of interest depicted in images represents particularly important information for the interpretation of their content. Spatial relations indeed carry rich semantics that are strongly tied with human perception. The research work presented in this thesis thus proposes to explore different generic approaches to the description of spatial information, in order to integrate them in high-level image recognition and understanding systems. First, we present an approach for the description of complex spatial configurations, where objects can be imbricated in each other. This notion is formalized by two novel spatial relations, namely enlacement and interlacement. We propose a model to describe and to visualize these configurations with directional granularity. This model is experimentally validated for applications in biomedical imaging, remote sensing and document image analysis. Then, we present a framework for learning composite spatial relations from image datasets. Inspired by bags of visual features approaches, this strategy allows to build vocabularies of spatial configurations occurring across images, at different scales. These structural features can notably be combined with local descriptions, leading to hybrid and complementary representations. Experimental results obtained for different datasets of structured images highlight the interest of this approach for image recognition and classification tasks