Sequential fuzzy cluster extraction and its robustness against noise

Partitional clustering methods such as C-Means classify all samples into clusters. Even a noise sample that is distant from any cluster is assigned to one of the clusters. Noise samples included in clusters bias the clustering result and tend to produce meaningless clusters. Our clustering method repeats to extract mutually close samples as a cluster and leave isolated noises unclustered. Thus, the produced clusters are less a ected by noises than those of C-Means. Because clusters can be obtained analytically by our method, repeated trials to avoid local minima are not necessary. The method is shown to be e ective for extracting straight lines from images in the experiments.