An experimental comparision of vector quantization techniques for Thai text image

Author

Tiawongsuwan, Manthana ; Lee, Wilaiporn

Author_Institution

Dept. of Electr. & Comput. Eng., King Mongkut´s Univ. of Technol. North Bangkok, Bangkok, Thailand

fYear

2014

Firstpage

64

Lastpage

67

Abstract

This paper presents the experimental comparison of vector quantization (VQ) methods for Thai text image. The compared methods include of Linde-Buzo-Gray (LBG), Enhanced LBG (ELBG) and Genetic algorithm (GA). The performance of VQ over gray image is measured with peak signal-to-noise ratio (PSNR) and processing time. The studied parameter are including of codebook number, cluster size, training image type, number of training image and noise. In addition, the results are shown in the verities aspects which are conducted to the appropriate approach in compression of Thai text image.

Keywords

genetic algorithms; image coding; vector quantisation; ELBG method; GA; Linde-Buzo-Gray method; Thai text image compression; VQ method; codebook number; enhanced LBG method; genetic algorithm; gray image; noise; peak signal-to-noise ratio; processing time; training image number; training image type; vector quantization technique; Genetic algorithms; Image coding; Image reconstruction; PSNR; Training; Vector quantization; Enhanced LBG (ELBG); Genetic algorithm (GA); Linde-Buzo-Gray (LBG); Thai Text Image; Vector Quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Communications (APCC), 2014 Asia-Pacific Conference on

Type

conf

DOI

10.1109/APCC.2014.7091606

Filename

7091606