TMW a New Method for Lossless Image Compression

We present a general purpose lossless greyscale image compression method TMW that is based on the use of linear predictors and implicit seg mentation In order to achieve competitive com pression the compression process is split into an analysis step and a coding step In the rst step a set of linear predictors and other parameters suitable for the image is calculated which is in cluded in the compressed le and subsequently used for the coding step This adaption allows TMW to perform well over a very wide range of im age types Other signi cant features of TMW are the use of a one parameter probability distribu tion probability calculations based on unquantized prediction values blending of multiple probability distributions instead of prediction values and im plicit image segmentation The method has applications beyond image com pression The work is also relevant to image seg mentation and image comparison For image compression the method has been compared to CALIC on a selection of test images and typically outperforms it by between and percent at the cost of considerably slower com pression In particular a bitrate of less than bpp has been achieved for the luminance band of the well known lenna image compared to bpp reported for CALIC in Wu