Arithmetic coding for lossless and loss-inducing image compression

Abstract

Summary form only given. Arithmetic coding has been applied to provide lossless and loss-inducing compression of optical, infrared, and synthetic aperture radar imagery of natural scenes. Several different contexts have been considered, including both predictive and nonpredictive variations, with both image-dependent and image-independent variations. In lossless coding experiments, arithmetic coding algorithms have been shown to outperform comparable variants of both Huffman and Lempel-Ziv-Welch coding algorithms by approximately 0.5 b/pixel. For image-dependent contexts constructed from high-order autoregressive predictors, arithmetic coding algorithms provide compression ratios as high as four. Contexts constructed from lower-order autoregressive predictors provide compression ratios nearly as great as those of the higher-order predictors with favorable computational trades. Compression performance variations have been shown to reflect the inherent sensor-dependent differences in the stochastic structure of the imagery. Arithmetic coding has also been demonstrated to be a valuable addition to loss-inducing compression techniques.<>