Abstract
In this paper we present an algorithm to further compress JPEG images without loss. The main contribution of this work is a sorting transform that allows reducing inter-block redundancy not exploited by the standard JPEG still image compression algorithm. Inter subband redundancy is also reduced by grouping of coefficients with similar statistics and entropy encoding in adaptive contexts. Tested on randomly selected JPEG images from the Internet, our approach reduces image file size on average by 14%.
Original language | English |
---|---|
Pages | 145-149 |
Number of pages | 5 |
State | Published - 2004 |
Event | Picture Coding Symposium 2004 - San Francisco, CA, United States Duration: 15 Dec 2004 → 17 Dec 2004 |
Conference
Conference | Picture Coding Symposium 2004 |
---|---|
Country/Territory | United States |
City | San Francisco, CA |
Period | 15/12/04 → 17/12/04 |