Comparison Between fuzzy C-Means Clustering (FCM) and geometrically guided condition Fuzzy C-Means clustering (ggc FCM)

Abstract

This paper compare between the traditional fuzzy C-Means clustering FCM and a proposed technique approach to geometrically guided fuzzy clustering. A modified fuzzy CMeans clustering (FCM), is extended to incorporate a priori geometrical information from spatial domain in order to improve image segmentation. This leads to a new algorithm where the cluster guidance is determined by the membership values on neighboring pixels. The algorithm of FCM is tested on synthetic and real image to demonstrate the improved image segmentation compared to traditional FCM.