A CONNECTIONIST APPROACH FOR CLUSTERING WITH APPLICATIONS IN IMAGE-ANALYSIS

A new neural network strategy for clustering is presented. The network works on the histogram and the process is similar to mode separation. The number of clusters are autonomously detected by the network and i t overcomes some major difficulties encountered by mode separation tec hniques. Clustering is done by first selecting the prototypes and then assigning patterns to one of the prototypes based on its distance fro m the prototype and the distribution of data. The network does not emp loy weight learning and is therefore faster than existing unsupervised learning networks. The network was applied to a wide class of problem s including gray level image reduction, color segmentation and remotel y sensed image segmentation. The experimental results obtained are pro mising.