Adaptive predictor for lossless image compression

A new method for lossless image compression of grey-level images is propose d. The image is treated as a set of stacked bit planes. The compressed vers ion of the image is represented by residuals of a non-linear local predicto r spanning the current bit plane as well as a few neighbouring ones. Predic tor configurations are grouped in pairs differing in one bit of the represe ntative point only. The frequency of predictor configurations is obtained f rom the input image. The predictor adapts automatically to the image, it is able to estimate the influence of neighbouring cells and thus copes even w ith complicated structure or fine texture. The residuals between the original and the predicted image are those that c orrespond to the less frequent predictor configurations. Efficiently coded residuals constitute the output image. To our knowledge, the performance of the proposed compression algorithm is comparable to the current state of t he art. Especially good results were obtained for binary images, grey-level cartoons and man-made drawings.