Segmentation d’images par FCM modifié considérant le contexte spatiale.

Abstract

In this work, we are interested by FCM [1] (fuzzy C- means) algorithm, one of the most popular clustering methods based on minimization of a criterion. However,the performance of this algorithm may significantly degrade in the presence of noise and intensity inhomogeneities. The use of algorithms which incorporate spatial information in their objective functions could improve the quality of segmentation. Solutions such as FCM_S [3] (FCM with Spatial constraints) can be used. Test on synthetic and IRM images show that these algorithms succeed to minimize the effect of noise and intensity inhomogeneities.