Lossless colour image compression using RCT for bi-level BWCA

Abstract

A novel lossless colour image compression scheme based on a reversible colour transform (RCT) and Burrows–Wheeler compression algorithm (BWCA) is presented. The lossless transformation from RGB to YUV colour space provides highly correlated pixel intensities in the transformed image, thus aiding in higher compression. The proposed scheme uses a two-pass Burrows–Wheeler transform (BWT) for the individual source image colour planes to enhance grey-level homogeneity in the 2-D space. Compression efficiency is compared against various schemes including the JPEG 2000 lossless compression scheme and the previously developed kernel move-to-front transform-based BWCA (kernel BWCA). Validation is carried out via small- and large-size images. The proposed method using RCT with bi-level BWT results in better compression by taking advantage of the redundancy in the grey levels brought by the YUV colour space. For small-size images, it achieves 45 and 126 per cent more compression than the JPEG2000 lossless and kernel BWCA scheme, respectively. Among the different schemes compared, the proposed scheme achieves overall best performance and is well suited to small- and large-size image data compression.