A Clustering Algorithm Based on Mathematical Morphology

Mathematical morphology is basically a set theory. It provides the concept of a structuring element to probe the image with arbitrary geometric patterns. A novel clustering algorithm based on mathematical morphology is presented. First, the data set is discretized, and then clusters are detected as well separated subsets by a hierarchical morphological operation procedure. An algorithm to determine connected components allows us to estimate the number of clusters. Experimental results demonstrate that the proposed clustering algorithm is able to cluster complex shaped data set better than the classical clustering algorithms, and find an optimal number of clusters