Title: Classified Vector Quantisation and population decoding for pattern recognition

Authors: Bailing Zhang, Sheng-Uei Guan

Addresses: Department of Computer Science and Software Engineering, Xi'an Jiaotong-Liverpool University, Suzhou, China. ' Department of Computer Science and Software Engineering, Xi'an Jiaotong-Liverpool University, Suzhou, China

Abstract: Learning Vector Quantisation (LVQ) is a method of applying the Vector Quantisation (VQ) to generate references for Nearest Neighbour (NN) classification. Though successful in many occasions, LVQ suffers from several shortcomings, especially the reference vectors are prone to diverge. In this paper, we propose a Classified Vector Quantisation (CVQ) to establish VQ for classification. By CVQ, each data category is represented by its own codebook, which can be implemented by some learning algorithms. In classification process, each codebook offers a generalised NN. The examples of handwritten digit recognition and offline signature verification are used to demonstrate the efficiency of the proposed scheme.

Keywords: LVQ; learning vector quantisation; classification; CVQ; classified vector quantisation; pattern recognition; population decoding; nearest neighbour; handwritten digit recognition; offline signature verification; handwriting.

DOI: 10.1504/IJAISC.2009.027294

International Journal of Artificial Intelligence and Soft Computing, 2009 Vol.1 No.2/3/4, pp.238 - 258

Published online: 19 Jul 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article