FUSION ET FOUILLE DE DONNEES GUIDEES PAR LES CONNAISSANCES : APPLICATION A L’ANALYSE D’IMAGE

Abstract

With the development of acquisition image techniques, more and more data coming from different sources of image become available. Multi-modality image fusion seeks to combine information from different images to obtain more inferences than can be derived from a single modality. In medical imaging based application fields, image fusion has emerged as a promising research area since the end of the last century. In this study we propose an architecture of an information fusion system based on the possibility theory for the segmentation of a target from multiple image sources. Our main application concerns the segmentation of multispectral MR images. The fusion process is decomposed into three basic phases. We model the information extracted from T2- weighted and PD-weighted images within a common framework, the retained formalism is the algorithm FPCM (Fuzzy Possibilistic C-means). We then combine this information with an operator of fusion taking into account redundancy and complementarities. We construct in the last phase a synthetic information allowing to exploit the results of fusion. Some results are presented and discussed