A Lossless Color Image Compression-Hierarchical Prediction & Context Adaptive Coding

Abstract

This paper introduces a algorithm of hierarchical prediction and context-adaptive arithmetic coding for the lossless compression of an RGB image. RGB image is first transformed to YCuCv by a reversible color transform(RCT). After that a conventional lossless image coder like CALIC is used to compress the luminance channel Y. The proposed method used for hierarchical scheme to encode the chrominance image. In this process, the chrominance image is decomposed by even row and odd row image. This decomposed image is compressed using arithmetic coding and decoded so that we can get the original image by using color reverse transformation after the image can be reconstructed and performance measure can be calculated. Hierarchical scheme enables the use of upper, left, and lower pixels for the pixel prediction, but the conventional raster scan prediction methods use upper and left pixels for pixel prediction.An appropriate context model for the prediction error is defined and the arithmetic coding is done to the error signal related to each context. It is shown that the proposed method i.e., hierarchical prediction and context-adaptive arithmetic coding further reduces the bit rates compared with JPEG2000 and JPEG-XR for several sets of images.