By Topic

VQ-agglomeration: a novel approach to clustering

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $33
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
J. -H. Wang ; Dept. of Electr. Eng., Nat. Taiwan Ocean Univ., Keelung, Taiwan ; J. -D. Rau

A novel approach called `VQ-agglomeration' capable of performing fast and autonomous clustering is presented. The approach involves a vector quantisation (VQ) process followed by an agglomeration algorithm that treats codewords as initial prototypes. Each codeword is associated with a gravisphere that has a well defined attraction radius. The agglomeration algorithm requires that each codeword be moved directly to the centroid of its neighbouring codewords. The movements of codewords in the feature space are synchronous, and will converge quickly to certain sets of concentric circles for which the centroids identify the resulting clusters. Unlike other techniques, such as the k-means and the fuzzy C-means, the proposed approach is free of the initial prototype problem and it does not need pre-specification of the number of clusters. Properties of the agglomeration algorithm are characterised and its convergence is proved

Published in:

IEE Proceedings - Vision, Image and Signal Processing  (Volume:148 ,  Issue: 1 )