Lossless Image Compression Based on Haar Multiresolution Representation

This paper presents a new algorithm that utilizes wavelet for lossless compression. Based on linear prediction and Haar multiresolution representation, the wavelet coefficients are divided into two parts according to their values. Those two parts and the location information of decomposing will be compressed effectively by arithmetic coding respectively. Experimental results show that the new algorithm is very practical especially for lossless compression.