Noisy Image Segmentation by a Robust Clustering Algorithm Based on DC Programming and DCA

We present a fast and robust algorithm for image segmentation problems via Fuzzy C-Means (FCM) clustering model. Our approach is based on DC (Difference of Convex functions) programming and DCA (DC Algorithms) that have been successfully applied in a lot of various fields of Applied Sciences, including Machine Learning. In an elegant way, the FCM model is reformulated as a DC program for which a very simple DCA scheme is investigated. For accelerating the DCA, an alternative FCM-DCA procedure is developed. Moreover, in the case of noisy images, we propose a new model that incorporates spatial information into the membership function for clustering. Experimental results on noisy images have illustrated the effectiveness of the proposed algorithm and its superiority with respect to the standard FCM algorithm in both running-time and quality of solutions.