The purpose of image compression is to reduce the bits used to represent the image depending on reduce the image redundancy. The methods of image compression could be summed up as the lossy and loseless compression. The lossy compression method leads to the information lost in some sense, so through this way, we could get the higher compression rate. Comparatively, the information lost is forbidden in the lossless compression, and we could not get high compression rate.In this article, we major in the application of the wavelet transform in the image compression. In pyramidal algorithm, an image is decomposed into multiresolution subbands with sets of tree-structured coefficients. A novel coefficient partition algorithm is introduced for splitting coefficients into two sets using a spatial orientation tree data structure. By splitting the coefficients, the overall theoretical entropy is reduced due to the different probability distribution for the two coefficient sets. A lossless coder based on the algorithm has a better performance than other wavelet-based lossless coders in our experiments, such as JPEG-2000 and S+P.
|