A wavelet coding preprocessing algorithm based on Bayesian estimation for image compression

Author

Jiang, Yun-Xiao ; Zhong, Zi-Fa ; Wang, Lun-Wen

Author_Institution

Inst. of Electron. Eng., Hefei, China

Volume

9

fYear

2005

fDate

18-21 Aug. 2005

Firstpage

5467

Abstract

This paper presents a wavelet coding preprocessing algorithm which adopts threshold values based on Bayesian estimation for image compression. The proposed algorithm considers a modification to the quantization stage of set partioning in hierarchical trees (SPIHT) (Said and Pearman, 1996), which achieves better peak signal to noise ratio (PSNR) and visual quality performance than existing embedded wavelet coders and JPEG. The performance of the proposed algorithm is demonstrated with extensive experimental results.

Keywords

Bayes methods; image coding; trees (mathematics); vector quantisation; wavelet transforms; Bayesian estimation; hierarchical trees; image compression; quantization; set partioning; signal to noise ratio; threshold value; visual quality; wavelet coding preprocessing; Bayesian methods; Codes; Frequency; Image coding; PSNR; Quantization; Signal processing algorithms; Transform coding; Wavelet coefficients; Wavelet transforms; Bayesian estimation; Wavelet coding; image compression;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on

Conference_Location

Guangzhou, China

Print_ISBN

0-7803-9091-1

Type

conf

DOI

10.1109/ICMLC.2005.1527910

Filename

1527910