The Enhanced LBG algorithm

Giuseppe Patanè and Marco Russo

Abstract

Clustering applications cover several fields such as audio and video data compression, pattern recognition, computer vision, medical image recognition, etc. In this paper we present a new clustering algorithm called Enhanced LBG (ELBG). It belongs to the hard and $K$-means vector quantization groups and derives directly from the simpler LBG. The basic idea we have developed is the concept of utility of a codeword, a powerful instrument to overcome one of the main drawbacks of clustering algorithms: generally, the results achieved are not good in the case of a bad choice of the initial codebook. We will present our experimental results showing that ELBG is able to find better codebooks than previous clustering techniques and the computational complexity is virtually the same as the simpler LBG.

Keywords: Clustering, Unsupervised Learning, LBG, GLA, LVQ, K-means, Hard c-means, Fuzzy c-means

[MyNNsELBG] G. Patanè and M. Russo, The Enhanced LBG Algorithm. Neural Networks, vol. 14 no. 9, pp 1219--1237, November 2001.

Download: [ps.gz] [pdf]