VQ Codebook Design Algorithm Based on Copula Estimation of Distribution Algorithm

Author

Guo, Xiaodong ; Wang, Lifang ; Zeng, Jianchao ; Zhang, Xueliang

Author_Institution

Complex Syst. & Comput. Intell. Lab., Taiyuan Univ. of Sci. & Technol., Taiyuan, China

fYear

2011

fDate

21-23 Nov. 2011

Firstpage

178

Lastpage

181

Abstract

The copula estimation of distribution algorithm is used in the work of codebook design, in which the selected population is the training vectors who coding with the concerned codeword. In each generation, the new codeword are generated by a copula and the estimated margins according to the selected population. The experimental results show that the proposed algorithm performs better than some classic codebook design algorithm such as LBG, ANT based algorithms, codebook-based tuba search algorithm and partitioning-based tuba search algorithm.

Keywords

search problems; vector quantisation; VQ codebook design algorithm; codebook-based tuba search algorithm; copula estimation; distribution algorithm; partitioning-based tuba search algorithm; training vectors; Algorithm design and analysis; Estimation; Partitioning algorithms; Signal processing algorithms; Training; Vector quantization; Vectors; Vector Quantizationt (VQ); codebook design; copula; estimation of distribution algorithm (EDA);

fLanguage

English

Publisher

ieee

Conference_Titel

Robot, Vision and Signal Processing (RVSP), 2011 First International Conference on

Conference_Location

Kaohsiung

Print_ISBN

978-1-4577-1881-6

Type

conf

DOI

10.1109/RVSP.2011.35

Filename

6114907