I. Introduction
Clustering is one of the fundamental human cognitive activities, and it has been used extensively in computer vision and pattern recognition [5], [2], [12], [14], [17], [28], [13], [19]. Of these methods, fuzzy clustering has been shown to be effective in solving the problem that each data point has a membership grade indicating its belongingness degree to each cluster, rather than assigning it to only one of the clusters [25], [27], [15], [11]. The fuzzy c-means (FCM) algorithm is one of the best known fuzzy clustering approaches, and its use for various applications is well described and analyzed in a previous paper by Bezdek (1992). This clustering method relies on the optimization of a specific cost function, and it works well when the clusters are compact or isotropic.