Abstract
A new mammography image segmentation method based on a perceptual and fuzzy approach is proposed. The main idea is to exploit some properties of the Human Visual System namely the directional and frequency selectivity and the fuzzy sets theory in order to segment the mammography images into meaningful components. A contrast enhancement is used in the case of suspicious cases identified during the segmentation process. The performance of the proposed method is objectively evaluated using the just-noticeable difference measure expressed through a saliency map of the mammography image before and after processing. The obtained results confirm the efficiency of the method in segmenting the images into meaningful regions.