Cette dernière décennie, la modélisation des villes 3D est devenue l'un des enjeux de la recherche multimédia et un axe important en reconnaissance d'objets. Dans cette thèse nous nous sommes intéressés à localiser différentes primitives, plus particulièrement les fenêtres, dans les façades de Paris. Dans un premier temps, nous présentons une analyse des façades et des différentes propriétés des fenêtres. Nous en déduisons et proposons ensuite un algorithme capable d'extraire automatiquement des hypothèses de fenêtres. Dans une deuxième partie, nous abordons l'extraction et la reconnaissance des primitives à l'aide d'appariement de graphes de contours. En effet une image de contours est lisible par l'oeil humain qui effectue un groupement perceptuel et distingue les entités présentes dans la scène. C'est ce mécanisme que nous avons cherché à reproduire. L'image est représentée sous la forme d'un graphe d'adjacence de segments de contours, valué par des informations d'orientation et de proximité des segments de contours. Pour la mise en correspondance inexacte des graphes, nous proposons plusieurs variantes d'une nouvelle similarité basée sur des ensembles de chemins tracés sur les graphes, capables d'effectuer les groupements des contours et robustes aux changements d'échelle. La similarité entre chemins prend en compte la similarité des ensembles de segments de contours et la similarité des régions définies par ces chemins. La sélection des images d'une base contenant un objet particulier s'effectue à l'aide d'un classifieur SVM ou kppv. La localisation des objets dans l'image utilise un système de vote à partir des chemins sélectionnés par l'algorithme d'appariement. / This last decade, modeling of 3D city became one of the challenges of multimedia search and an important focus in object recognition. In this thesis we are interested to locate various primitive, especially the windows, in the facades of Paris. At first, we present an analysis of the facades and windows properties. Then we propose an algorithm able to extract automatically window candidates. In a second part, we discuss about extraction and recognition primitives using graph matching of contours. Indeed an image of contours is readable by the human eye, which uses perceptual grouping and makes distinction between entities present in the scene. It is this mechanism that we have tried to replicate. The image is represented as a graph of adjacency of segments of contours, valued by information orientation and proximity to edge segments. For the inexact matching of graphs, we propose several variants of a new similarity based on sets of paths, able to group several contours and robust to scale changes. The similarity between paths takes into account the similarity of sets of segments of contours and the similarity of the regions defined by these paths. The selection of images from a database containing a particular object is done using a KNN or SVM classifier.

[pt] Esta dissertação apresenta novos algoritmos aproximados e uma abordagem exata para a resolução de um problema de correspondência inexata de grafos. O problema considerado é o de correspondência entre um grafo representando um modelo genérico e outro representando dados a serem reconhecidos. Assumi-se que o grafo dos dados possui mais vértices que o do modelo. A motivação para o estudo desse problema vem de problemas de reconhecimento de cenas, que consistem na caracterização dos objetos envolvidos em uma determinada cena, assim como das relações existentes entre eles. Uma aplicação para este problema na área de reconhecimento de imagens médicas é a de efetuar-se o reconhecimento de estruturas 3D do cérebro humano, a partir de imagens obtidas por ressonância magnética. Tais imagens são previamente processadas por algum método de segmentação automática e o processo de reconhecimento consiste na busca da correspondência estrutural entre a imagem e um modelo genérico, tipicamente definido como um atlas de imagens médicas. Foram propostos novos algoritmos aproximados, tais como um algoritmo construtivo guloso aleatorizado, um procedimento de reconexão de caminhos e um GRASP que combina estes com uma técnica de busca local. Além disso, foi proposta uma formulação original do problema como um problema de programação linear inteira, que permitiu a resolução de algumas instâncias de forma exata. / [en] This dissertation presents new approximation algorithms and an exact approach to the solution of an inexact graph matching problem. The problem consists in finding the best match between a generic model graph and a graph representing an image, the latter with more nodes than the former. The motivation for studying this problem comes from a scene recognition problem, which consists in characterizing objects involved in a given scene and the relationships between them. An application of this problem appears in the analysis of medical images and consists in recognizing 3-dimensional structures in the human brain using images obtained by magnetic resonance. Such images must be previously processed by an automatic segmentation method and the recognition process consists in the search of an structural matching between the image and a generic model, typically defined as an atlas of medical images. New heuristics are proposed, such as a greedy randomized construction algorithm, a path relinking procedure and a GRASP heuristic that combines them with a local search technique. Furthermore, an original integer formulation of the problem based on integer multicommodity flows is proposed, which makes possible the exact solution of medium- sized instances.