題名: An Improved Full-Search Scheme for the Vector Quantization Algorithm Based on Triangle Inequality
作者: Lu, T. C.;Chang, C. C.
關鍵詞: Vector quantization;Euclidean distance;Triangle inequality;Full search
日期: 2008-07
上傳時間: 2009-12-17T06:57:44Z
出版者: Asia University
摘要: In order to speed up the computation time of searching the closest codeword in VQ compression algorithm, Mielikainen proposed an efficient full-search scheme
based upon the law of cosines. However, some computational redundancies still exist in
Mielikainen?s scheme. Hence, Pan et al. developed a scheme based upon triangle projection
to improve Mielikainen?s scheme. In this paper, we introduce a scheme based upon
triangle inequality to speed up the encoding time of the VQ algorithm. The experimental
results show that the proposed method can significantly decrease the computation time
and maintain the same coding quality as the full-search VQ algorithm.
Furthermore, the
memory storage used in the proposed scheme is less than those in Mielikainen?
s and Pan
et al.?s schemes.