Image Compression Technology Based On All-Phase Digital Filtering

As developing of the society, the requirements of information become more andmore large. Image information has already been used in the fields of Digital TV, VideoTelephone, etc. How to solve the contradiction existing between the imageinformation and the accommodation of channel is the most important issue beingconsidered. The image compressing technology based on all-phase subband filter is anew compressing algorithm, and has a lot of advantages compared with the otheralgorithms basing on some conventional transforms. Combining with the algorithm ofimage characteristic extracting, this algorithm can get a good quality of reconstructingimage at the same time having a considerable compressing ratio. In this scheme, the primitive image is decomposed into four subbands usingall-phase digital filter basing on DCT transform. Only the low-frequency and thesecond-frequency subband can be retained. To the low-frequency subband, we usesubsampling and DPCM encoding. And to the second-frequency subband, we useimmersion algorithm to extract the watersheds and watercourses of image, and encodethe direction of these lines using Freeman encoding. Finally, all data are encoded byadaptive arithmetic encoding. During decoding, we use the method of convexprojection. The experiments' results show that this algorithm has a good speed ofcalculation, a simple method of designing filter, high compressing ratio and a goodquality recovering image.