Lossless Image Compression Using List Update Algorithms

摘要: We consider lossless image compression using a technique similar to bZip2 for sequential data. Given an represented with matrix of pixel values, we different approaches linearising the into sequence and then encoding Move-To-Front list update algorithm. In both linearisation stages, exploit locality present in images achieve encodings that are as compressed possible. few approaches, particular Hilbert space-filling curves, image. Using natural model introduced by Albers et al. [J. Comput. Syst. Sci. 2015], establish advantage curves over other techniques such row-major or column-major preserving during linearisation. also use result Angelopoulos Schweitzer ACM 2013] select best algorithm linearised sequence. summary, our theoretical results show combination has approaches. verify this experiments on dataset consisting categories images.