WorldCat Identities

École polytechnique universitaire (Tours)

Overview
Works: 533 works in 605 publications in 2 languages and 652 library holdings
Genres: Academic theses 
Roles: Degree grantor, Other, Editor, Organizer of meeting
Publication Timeline
.
Most widely held works by École polytechnique universitaire (Tours)
Détection d'évènements dans une séquence vidéo by Sébastien Lefèvre( Book )

2 editions published in 2002 in French and held by 4 WorldCat member libraries worldwide

Here we are concerned by multimedia data indexing by use of pertinent sample searching. Our work focuses on video sequence analysis in order to detect some predefined events. This search being contextual, we propose an architecture and some rapid tools, both generic, for building specific indexing systems. We deal in particular with the following problems : data temporal segmentation, background and object separation, background structure extraction, object tracking (rigid or non-rigid objects, with or without learning) and audio data analysis. In order to solve these different problems, the generic tools we are proposing are based on semi-local analyses, multiresolution approaches, hidden Markov models, and the active contour method. The architecture and the tools introduced here have been validated through several applications
Identification et résolution de problèmes de sécurité dans un habitat domotisé by Ali Larab( Book )

2 editions published in 2006 in French and held by 4 WorldCat member libraries worldwide

Les réseaux domestiques tels qu'ils sont conçus actuellement sont plus vulnérables que les réseaux informatiques ordinaires. Cela est dû aux caractéristiques intrinsèques de ces réseaux et de leurs utilisateurs. Pour les sécuriser, les outils de sécurité classiques ne sont pas toujours adaptés. Soit ils ne peuvent pas être appliqués sur de tels réseaux, soit ils sont trop contraignants pour leurs utilisateurs. Nous avons alors proposé dans ce rapport des solutions qui comblent les vides laissés par les solutions de sécurité classiques et par celles proposées dans la littérature. En plus des solutions existantes applicables aux réseaux domestiques, nous en avons proposé d'autres pour sécuriser les communications avec les utilisateurs distants (nomades), mais aussi pour sécuriser les communications internes
Contribution à la résolution des problèmes d'ordonnancement de type monogamme, multimachine (Flow-Shop Hybride) by Antony Vignier( Book )

1 edition published in 1997 in French and held by 4 WorldCat member libraries worldwide

Ce travail s'intéresse aux problèmes qui se posent dans les ateliers production en ligne et plus précisément sur ceux ou les gammes de fabrication de produits fabriqués sont identiques pour tous les produits (Flow-Shop). De plus, une opération de la gamme peut être réalisée par une ou plusieurs machines à chaque étage. L'atelier comporte K étages. En vue d'une plus grande efficacité d'intervention, une notation est proposée. L'état de l'art permet de mettre en évidence le peu de problèmes actuellement résolus et surtout l'ensemble de ceux qui restent à traiter des méthodes de résolution sont proposées pour résoudre différents problèmes prenant en compte des contraintes et de critères divers. Enfin, une plateforme d'aide à la construction progressive de procédures par séparation et évaluation et de tests (PCPSET) pour résoudre les problèmes de Flow-Shops Hybrides
Analyse de signaux vidéos et sonores : application à l'étude de signaux médicaux by Emmanuel Dellandréa( Book )

2 editions published in 2003 in French and held by 4 WorldCat member libraries worldwide

The work deals with the study of multimedia sequences containing images and sounds. The analysis of images sequences consists in the tracking of moving objects in order to allow the study of their properties. The investigations have to enable the understanding of sounds when correlated to events in the image sequence. One generic method, based on the combination of regions and contours tracking, and one method adapted to homogeneous objects, based on level set theory, are proposed. The analysis of audio data consists in the development of an identification system based on the study of the structure of signals thanks to their coding and Zipf laws modeling. These methods have been evaluated on medical sequences within the framework of the gastro-oesophageal reflux pathology study, in collaboration with the Acoustique et Motricité Digestive research team of the University of Tours
Les contours actifs, une méthode de ségmentation : application à l'imagerie médicale by Jean-Jacques Rousselle( Book )

2 editions published in 2003 in French and held by 4 WorldCat member libraries worldwide

The segmentation methods of images are numerous ; all have advantages but do not give full satisfaction. All must be adapted according to the application which has to be carried out. Active contours or deformable models made it possible to avoid to chain the contour points but require the adjustment of many parameters. Active contours that we have studied are implemented using a greedy algorithm. First, we propose an alternative based on a minimization by genetic algorithm.Then we propose three approaches to regulate the parameters which control the evolution of contour. Design of experiments makes it possible from a set of images to very quickly choose a set of powerful parameters. The genetic algorithms can be used to optimize the parameters. Finally we propose an original approach where the parameters are local and randomly defined. These autonomous snake allow an evolution of contours without any adjustment. The applications use various images, but in particular medical images
Algorithmes hybrides d'apprentissage de chaines de Markov cachées : conception et applications à la reconnaissance des formes by Thierry Brouard( Book )

1 edition published in 1999 in French and held by 3 WorldCat member libraries worldwide

The main point of this work is based on the quality of modelization of data (called observations) made by hidden Markov models (HMMs). Our goal is to propose algorithms that improve this quality. The criterion used to quantify the quality of HMM is the probability that a given model generates a given observation. To solve this problem, we use a genetic hybridization of HMM. Using genetic algorithms (GAs) jointly to HMM permits two things. First, GAs let us to explore more efficiently the set of models, avoiding local optima. Second, GAs optimize an important characteristic of HMM : its number of hidden states. The most efficient hybrid algorithm finds the best HMM for a given problem, by itself. This means that the GA designs a set of states and the associated transition probabilities. Many explications have been done in the framework of this thesis, in many domains like image recognition, time series prediction, unsupervised image segmentation and object tracking in sequences of images. The new algorithms proposed here are appliable to all domains (peovided that hypothesis related to HMM are satisfied). They allow a fast and efficient training of HMM, and an entirely automatic determination of the architecture (number of states, transition probabilities) of the HMM
Projet HM2PH, génération automatique de plans et visite virtuelle d'habitats adaptés pour personnes handicapées by Arnaud Puret( Book )

2 editions published in 2007 in French and held by 3 WorldCat member libraries worldwide

This study was realized in the framework of the HM2PH project, in the HaNT team of the LI of Tours. The design of an adapted house for a disabled person requires an interior arrangement according to the capacities, the disabilities and the wishes of this person. We have developed a house layout generator to automate the respect of the disability and building constraints. The topological research and the geometrical optimisation of the layouts are done with constraint satisfaction methods. Clustering and visualization methods (circular display for the K-Means, NicheWorks method for the HAC) are used to display the layouts to the user. Then, a 3D virtual visit of the layout is generated. The user can act on the furniture which has its working reproduced. The interactive interface to control the house automation system, realised with the virtual visit tool, is a first step to a virtual environment server
Un nouvel algorithme de boosting pour les réseaux de neurones récurrents : application au traitement des données sequentielles by Mohammad Assaad( Book )

2 editions published in 2006 in French and held by 3 WorldCat member libraries worldwide

The work of this thesis deals with the proposal of a new boosting algorithm dedicated to the problem of learning time-dependencies for the time series prediction, using recurrent neural networks as regressors. This algorithm is based on the boosting algorith and allows concentrating the training on difficult examples. A new parameter is introduced to regulate the influence of boosting. To evaluate our algorithm, systematic experiments were carried out on two types of problems of time series prediction : single-step ahead predicton and multi-step ahead prediction. The results obtained from several series of reference are close to the best results reported in the literature
Enumération en ordonnancement multicritère : application à un problème bicritère à machines parallèles by Karima Bouibede-Hocine( Book )

2 editions published in 2007 in French and held by 3 WorldCat member libraries worldwide

Numerous scheduling problems were the subject of many studies in the literature. But the majority of them are related to the solutions of problems with a single criterion, although real life is quite different since problems often involves multiple criteria. The solution of this kind of problem consists in finding a set of points corresponding to the best tradeoffs between different criteria. These points are called Pareto Optima. Numerous solutions methods are present in the literature to solve multicriteria scheduling problems. Some of these methods are based on the determination of only one Pareto optimum. Others are based on the enumeration of all the Preto optima. A synthesis of these methods is presented in [T4Kindt and Billaut 2006]. In this PhD thesis we start by tackling an important point of combinatorial optimisation : that of complexity theory when dealing with the counting or enumeration of pareto optima in the case of multicriteria optimisation. A synthesis of this part was proposed in [T'Kindt et al., 2005]. Then we focus on the comparison of threee methods of enumeration of strict Pareto Optima for the uniform parallel machines scheduling problem, Q|ri,di,Cmax,Lmax. The first is based on the e-constraint approach ([Klein and Hannan, 1982]), the seconde method is the two phases method which was initialy proposed by [ulungu and Teghem, 1995] to solve a bicriteria knapsack problem. The last method is the branch, bound and delay procedure which is based on the exploration of a search tree whose nodes are built only once
Sustainability in heritage protected areas : [European Urban Summer School 2014] by Laura Verdelli( Book )

2 editions published in 2015 in French and English and held by 3 WorldCat member libraries worldwide

The present book contains the proceedings of the fifth European Urban Summer School (EUSS) which took place in Tours, France, from 1st - 8th September 2014, organised by the École Polytechnique de l'Université de Tours, Département Aménagement et Environnement (EPU-DAE). The theme.is 'Heritage conservation and sustainable urban development'. It is the third and last of those including the papers of YPPA winners and sponsored by mI&M. The YPPA papers are those from the three winners: Fernando Navarro Carmona - elCASC - from Spain, Cexiang Foo from Singapore and Nasos Alexis from Crete, Greece, and from the runner-up Anna Peralta Zaragoza, also from Spain. Responsible for the 5th EUSS and for editing this publication is Laura Verdelli. The partner organisations are indebted to her for her huge efforts to make both the Summer School and this book a success. Tours has proved itself a worthy case study for examining the theme of heritage and sustainability as this book shows. We hope you enjoy and get some new ideas from reading its content. The 5th EUSS / 3rd YPPA have once again confirmed that a few days of intensive interaction, hard work and fun can produce many useful new ideas from, and friendships amongst, young planning professionals and tutors from diverse countries who participated and have contributed to this book
Contribution des réseaux de Petri à l'étude de problèmes de recherche opérationnelle by Pascal Richard( Book )

1 edition published in 1997 in French and held by 3 WorldCat member libraries worldwide

Un réseau de Petri est un modèle mathématique utilisé pour modéliser et analyser les systèmes à événements discrets. Nous présentons dans ce mémoire leur utilisation pour des problèmes de recherche opérationnelle. L'état de l'art montre l'intérêt de cet outil pour aborder cette problématique. L'étroit lien entre la programmation mathématique et les réseaux de Petri est ensuite mis en évidence. Précisément, nous montrons que le problème de programmation mathématique se réduit polynomialement à un problème d'ordonnancement dans les réseaux de Petri. Ce résultat repose sur une nouvelle condition nécessaire et suffisante d'accessibilité pour l'équation des marquages. Nous présentons enfin une application industrielle de ces résultats dans le cadre d'un logiciel d'aide à la décision pour la planification de production dans l'industrie du verre
Sous-automates à nombre fini d'états : application à la compression de dictionnaires électroniques by Lamia Tounsi( Book )

2 editions published in 2007 in French and held by 3 WorldCat member libraries worldwide

This work deals with the study of internal structure of acyclic automaton. Thus, we propose an algorithm to compute all subautomata of a given automaton. This study can be used in applications whose aim is to decompose a very large FSA into smaller ones, to discover frequently occurring data and to reduce memory consumption. The second part of our work is devoted to the application of our algorithm for compression and indexing of automata that represent electronic dictionaries. Also, we propose a compression algorithm to reduce the memory required to store the automata and to preserve an effective access to data. The main propositions are, on the one hand, the application of the direct acyclic word graph, initially dedicated for indexing text, to index the subautomata, and, on the other hand, heuristic to select the most interesting substructure to factorize. The best candidates to factorize must increase memory storage efficiency and reduce the size of the initial automaton
Authentification par signature manuscrite sur support nomade by Matthieu Wirotius( Book )

2 editions published in 2005 in French and held by 3 WorldCat member libraries worldwide

Le travail présenté dans cette thèse concerne l'authentification par signature manuscrite en ligne. Tout d'abord, nous avons étudié la caractérisation des signatures par extraction et sélection de caractéristiques globales. Les caractéristiques innovantes proposées reposent sur l'étude de la fractalité et sur une étude multi-résolution des signatures manuscrites en ligne. Ensuite, dans le but d'améliorer les résultats obtenus en utilisant une méthode appelée Dynamic Time Warping, nous présentons des modifications pour la phase de mise en correspondance des points des signatures ainsi que de nouvelles méthodes de calcul de la distance entre signatures utilisant l'ensemble des informations disponibles. Nous avons finalement construit une architecture tenent compte des contraintes liées à la volonté de stocker les données relatives à la signature dans une carte à puce. Les tests que nous avons réalisés sur des bases réelles de signatures montrent que les choix effectués étaient judicieux
Extraction et formalisation de la sémantique des liens hypertextes dans des documents culturels, scientifiques et techniques by Moustafa Al-Hajj( Book )

2 editions published in 2007 in French and held by 3 WorldCat member libraries worldwide

The use of hypertext links on the web makes sites more attractive and easier to read and allows enrichment of sites by information coming from other sites. However, this links produce some difficulties for readers and search engines. The hypertext links are carrying semantic information which, if it were completely formalized, would be exploitable by programs to improve navigation and research of information, and would take its place in the emergence of semantic web. In this thesis, we propose an original methodology for the formal semantic extraction of hypertext links. The suggested method has been tested on the links of a corpus. The formalism RDF has been used to represent the link semantics. Ontology for the links specific to the field of biographies of famous people was made up starting from the link semantics extracted and then represented in RDFS. Some tools of supervised learning and of web pages characterization by keywords has been used to help with the formal extraction of semantics
Une méthode pour l'évolution de schémas XML préservant la validité des documents by Denio Duarte( Book )

2 editions published in 2005 in French and held by 3 WorldCat member libraries worldwide

We propose an approach to support adminstrators of XML applications, which allows XML schema evolution while preserving the consistency of existing data without any need of their modification. The user provides the system with a new document D' (possibly invalid with respect to the present schema) resulting from updating a valid document D, that triggers the evolution of the schema. From D', the system builds candidate schemas that can replace the original one and that preserve the data consistency. The system uses the associated tree automaton to find the positions in the schema that should be updated. The schema evolution is implemented by an algorithm (named GREC) that performs changes on the graph of a finite state automaton and that generates regular expressions for the modified graphs. Each regular expression proposed by GREC is given as a choice of schema to the administrator
Problèmes d'ordonnancement dans un flowshop hybride avec recirculation sous contraintes de gestion du personnel by Sylvain Bertel( Book )

2 editions published in 2001 in French and held by 3 WorldCat member libraries worldwide

This work has been achieved within the context of the research collaboration with an industry (CIFRE), the customer management service Atos-Origin and the computer science laboratory of the university of Tours (LI). In this document, we propose in a first time a survey on hybrid flowshop schedulling problems that minimize the wheighted number of late jobs and manpower scheduling problem. A scheduling problem from an industrial context of cheque processing is studied. For this problem, two heuristics based on priority rules and a metaheuristic are proposed. From developed heuristics, we describe an integrated procedure which can be used at different decision levels. Regarding manpower scheduling, we propose linear programming model and a heuristic that gives the size of the workshop. We notice that a planning defined at a tactical level is moire efficient than a planning defined at an operational level. We end this document with the presentation of a software called PLANIF'00
Algorithmes approchés pour des problèmes d'ordonnancement multicritères de type job shop flexible et job shop multiressource by Geoffrey Vilcot( Book )

2 editions published in 2007 in French and held by 3 WorldCat member libraries worldwide

This thesis was realized in the context of an industrial collaboration. In this work, we studied the multicriteria flexible job shop scheduling problem and the multicriteria multiprocessor tasks job shop scheduling problem. Our purpose was to find an approximation of the pareto frontier for these problems. Different meta heuristics were developped, such as Tabu search algorithms and genetic algorithms. Different versions of our methods were proposed for the two considered problems. Experiments show the good performances of our algorithms, both qualitative and computational aspects of our methods show satisfying results. This thesis was done in collaboration with Volume Software company within the framework of the development of the "DirectPlanning" scheduling engine
Contributions aux modèles de Markov cachés : métaheuristiques d'apprentissage, nouveaux modèles et visualisation de dissimilarité by Sébastien Aupetit( Book )

2 editions published in 2005 in French and held by 3 WorldCat member libraries worldwide

In this PhD thesis, we present many contributions aimed at the improvement on the utilization of hidden Markov Models (HMMs) in artificial intelligence systems. We considered three main objectives : improving HMM training, experimenting a new HMM and visualizing interaction among HMMs. In the first part, we propose, evaluate and compare many new adaptations of classic biomimetic metaheuristics (genetic algorithms, API artificial ants algorithm, particle swarm optimization) applied to the problem of HMMs training. In the second part, we propose a new kind of HMM which we named symbols substitution hidden Markov models (SSHMMs). A SSHMM allows an expert to incorporate \textit {a priori} knowledge in the training and the recognition tasks. First experiments with such models show that SSHMMs are of great interests at least for images learning and recognition. In the third part, we propose a new visualization technique to tackle dissimilarity. This technique, which we named the pseudo-euclidean scatterplots matrix (PESM), allows a better understanding of interaction between HMMs. This PESM is built from techniques which we named indefinite kernel principal component analysis (IKPCA). Finally, our research concludes with the description of the HMMTK software library developed along this work. The library integrates parallelization mechanisms and algorithms developed during the thesis
Ségmentation de maillages 3D à l'aide de méthodes basées sur la ligne de partage des eaux by Sébastien Delest( Book )

2 editions published in 2007 in French and held by 3 WorldCat member libraries worldwide

Mesh segmentation is a necessary tool for many applications. The mesh is decomposed into several regions from surface or shape information. In the last several years, many algorithms have been proposed in this growing area, with applications in many different areas as 3D shape matching and retrieval, compression, metamorphosis, collision detection, texture mapping, simplification, etc. First, we propose a review of mesh segmentation methods. We discuss about the algorithms relative to the two main types of methods: the patch-type segmentation and the part-type segmentation. We focused on the watershed transformation and proposed new approches relativing to the height function and strategies to avoid over segmentation produced by the watershed
Une méthode d'analyse de données textuelles pour les sciences sociales basée sur l'évolution des textes by Hubert Marteau( Book )

2 editions published in 2005 in French and held by 3 WorldCat member libraries worldwide

This PhD Thesis aims at bringing to sociologists a data-processing tool wich allows them to analyse of semi-directing open talks. The proposed tool performs in two steps : an indexation of the talks followed by a classification. Usually, indexing methods rely on a general stastistical analysis. Such methods are suited for texts having contents and structure ( literary texts, scientific texts,...). These texts have more vocabulary and structure than talks (limitation to 1000 words for suche texts). On the basis of the assumption that the sociological membership strongly induces the form of the speech, we propose various methods to evaluate the structure and the evolution of the texts. The methods attempt to find new representations of texts (image, signal) and to extract values from these new representations. Selected classification is a classification by trees (NJ). It has a low complexity and it respects distances, then this method is a good solution to provide a help to classification
 
moreShow More Titles
fewerShow Fewer Titles
Audience Level
0
Audience Level
1
  Kids General Special  
Audience level: 0.90 (from 0.86 for Contributi ... to 0.93 for Une métho ...)

Alternative Names
Ecole polytechnique de l'université de Tours

EPU

EPU (Tours)

Polytech'Tours

Université de Tours. Ecole polytechnique universitaire

Université de Tours. EPU

Université François Rabelais Tours Ecole polytechnique

图尔大学综合理工学院

图尔工程师学院

Languages
French (36)

English (1)