DocumentCode
1582353
Title
Weibull based incremental adaptive arithmetic coding with file partition technique
Author
Wiengpon, A. ; Surarerks, A.
Author_Institution
Dept. of Comput. Eng., Chulalongkorn Univ., Bangkok, Thailand
fYear
2010
Firstpage
1068
Lastpage
1073
Abstract
In this paper, an improvement of incremental adaptive arithmetic coding and the probability estimation of symbols are proposed. Experiments are designed to observe the improvement in compression ratio (CR) of English text file of three different techniques. The probability density function (PDF) of Weibull distribution is firstly introduced to estimate the initial probability of input symbols. The best elementary result for improvement of CR illustrates that the shape and scale parameter (α, β) should be 0.3 and 6.4 respectively. In addition, eliminating unused symbol and file partition algorithms are also applied. The performance evaluation of CR was conducted with nine ordinary English novels. CR value increases positively comparing to the result from incremental adaptive arithmetic coding (IAAC1) in every experimental files with the improvement percentage range of (0.0785 - 1.0288%).
Keywords
Weibull distribution; adaptive codes; arithmetic codes; data compression; natural language processing; text analysis; English text file; Weibull distribution; compression ratio; file partition technique; incremental adaptive arithmetic coding; performance evaluation; probability density function; probability estimation; Encoding; Estimation; Image coding; Partitioning algorithms; Shape; Weibull distribution;
fLanguage
English
Publisher
ieee
Conference_Titel
Communications and Information Technologies (ISCIT), 2010 International Symposium on
Conference_Location
Tokyo
Print_ISBN
978-1-4244-7007-5
Electronic_ISBN
978-1-4244-7009-9
Type
conf
DOI
10.1109/ISCIT.2010.5665145
Filename
5665145
Link To Document