Title :
Image coding using fuzzy vector quantizer
Author :
Li, Haizhou ; Jin, Lianwen
Author_Institution :
Inst. of Radio & Autom., South China Univ. of Technol., Guangzhou, China
Abstract :
The fuzzy vector quantizer (FVQ) technique is applied to image coding when a square image subblock is mapped into a reproduction codeword. The FVQ is also tested on a database of Gauss Markov sources to demonstrate its performance in coding highly correlated data. An attempt is made to determine whether the use of softer decisions in the FVQ training step would optimize the codeword distribution and make the quantizer adapt well to random initial codebooks. The experiments show that FVQ coding performs better than standard methods
Keywords :
fuzzy set theory; image coding; vector quantisation; Gauss Markov sources; fuzzy vector quantizer; highly correlated data; image coding; reproduction codeword; softer decisions; square image subblock; Automation; Channel capacity; Clustering algorithms; Code standards; Digital images; Image coding; Partitioning algorithms; Pixel; TV; Testing;
Conference_Titel :
TENCON '92. ''Technology Enabling Tomorrow : Computers, Communications and Automation towards the 21st Century.' 1992 IEEE Region 10 International Conference.
Conference_Location :
Melbourne, Vic.
Print_ISBN :
0-7803-0849-2
DOI :
10.1109/TENCON.1992.271876
