The pseudo-distance technique for parallel lossless compression of color-mapped images. (August 2015)

• Record Type: Journal Article
• Authors: Koc, Basar; Arnavut, Ziya; Koçak, Hüseyin
• Is Part Of: Computers & electrical engineering. Volume 46(2015)
• Page Start: 456
• Page End: 470
• Publication Date: 2015-08
• Subjects: Parallel image compression -- Hyper-Threading -- Color-mapped images -- PDT -- BWT -- PNG; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2015.01.003
• Languages: English
• ISSNs: 0045-7906

Abstract

Graphical abstract: Highlights: A pseudo-distance technique (PDT2) for lossless compression of images is given. Parallelization of the technique is presented. The performance of the PDT2 in Hyper-Threading is evaluated. Abstract: Data compression is a challenging process with important practical applications. Specialized techniques for lossy and lossless data compression have been the subject of numerous investigations during last several decades. Previously, we studied the use of the pseudo-distance technique (PDT) in lossless compression of color-mapped images and its parallel implementation. In this paper we present a new technique (PDT2) to improve compression gain of PDT. We also present a parallelized implementation of the new technique, which results in substantial gains in compression time while providing the desired compression efficiency. We demonstrate that on non-dithered images PDT2 outperforms PDT by 22.4% and PNG by 29.3%. On dithered images, PDT2 achieves compression gains of 7.1% over PDT and 23.8% over PNG. We also show that the parallel implementation of PDT2, while compromising compression less than 0.3%, achieves near linear speedup and utilization of Intel Hyper-Threading technology on supported systems improves speedup on average 18%.