An algorithmic study on context modelling for lossless image compression

Statistical context modelling is a powerful and versatile technique for los sless image coding. A key issue in context modelling is how to increase the order of model without drastically increasing the model cost. We take an a lgorithmic approach to address the issue and propose a few heuristical opti mization techniques that fine tune models of relatively few parameters in t he sense of entropy minimization. Our experimental results indicate that th ese techniques are quite effective and achieve the lowest lossless bit rate s so far over a variety of test images.