An efficient fuzzy c-means clustering algorithm for image data
by
Yung-Sheng Chen
Department of Electrical Engineering, Yuan Ze University
Coauthors: Bor-Tow Chen

The clustering process may be quite slow when there is a large data set to be clustered. In this paper, an efficient fuzzy c-means clustering method qFCM, based on the quad tree application to multi-spectral image feature compression and/or an aggregation process to reduce the number of exemplars, is investigated for image analysis. A set of fourteen images is used for experiments, comparison and discussion. Performances are reported by the mean reduction rate, speedup, mean correspondence rate, and root mean square error. Results show that on the measure of mean reduction rate, our algorithm can reach 98% above. Average speedups of as much as 40~200 times a traditional implementation FCM are obtained using the proposed algorithm, while producing partitions that are equivalent to those produced by FCM. On the measure of root mean square error, the proposed algorithm is the better one indicated in the experiment of clustering a noisy image.

Date received: March 19, 2004

Copyright © 2004 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # canw-15.