I. Introduction
Clustering is an important data processing method, which has been widely applied in the areas of pattern recognition [1], image processing [2], data mining [3], etc. The aim of clustering is to partition datasets into meaningful groups with similar samples. Fuzzy sets proposed by Zadeh [4] were introduced into the clustering [5]. So far, fuzzy clustering has been widely studied and applied in variety of substantive domains [6]–[10]. Bezdek et al. [11] proposed a fuzzy c-means clustering algorithm (FCM) that has become the most well-known fuzzy clustering algorithms. The FCM algorithm introduces the fuzziness into the belongingness of each sample. Each of the samples is assigned a membership grade from 0 to 1. The objective function of the classical FCM algorithm is defined by fuzzy memberships and Euclidean distances of samples to cluster centers.