Determining number of clusters and prototype locations via multi-scale clustering

In clustering algorithms, it is usually assumed that the number of clusters is known or given. In the absence of such a priori information, a procedur e is needed to find an appropriate number of clusters. This paper presents a clustering algorithm that incorporates a mechanism for finding the approp riate number of clusters as well as the locations of cluster prototypes. Th is algorithm, called multi-scale clustering, is based on scale-space theory by considering that any prominent data structure ought to survive over man y scales. The number of clusters as well as the locations of cluster protot ypes are found in an objective manner by defining and using lifetime and dr ift speed clustering criteria. The outcome of this algorithm does not depen d on the initial prototype locations that affect the outcome of many cluste ring algorithms. As an application of this algorithm, it is used to enhance the Hough transform technique. (C) 1998 Elsevier Science B.V. All rights r eserved.