Compression Based On Wavelet Transform Image Coding Techniques

In recent years, with the rapid development of multimedia technology of the computer and network communication, the demands of people for the image and multimedia information are increasing constantly. Unprocessed image data is enormous, the image transmission, processing and store are limited. So to study high-efficient image data compress coding method---how to deal with, to organize image data, is more important in application field, image data compress coding technology has become one key technology in multimedia and communication field already.The wavelet analysis, a new subject developed ten years ago, has demonstrated the good prospect in the image compression aspect. It has abandoned the premise supposition in traditional FFT----stability, it overcome some weakness of traditional FFT, deals with thickly to the smooth part of the image; deal with meticulously to the detail part. Meanwhile, it can resolve the image on much scale, to do multi-resolution analysis more. In this way, according to sub difference of characteristic of image of every scale, we can carry on the different treatment. The image rebuild with this method accord with the subjective characteristic of the human eye, having dispelled a block effect completely, the code efficiency of the image is improved greatly too.This article first discuss the essentiality and feasibility of image compress, introduced the elementary theory of wavelet transform and the concept of MRA, as well as the fast algorithm of wavelet decomposition and construction ---Mallat algorithm, and has summarized to the traditional classics image coding technology.Has carried on the discussion and the analysis to image compression characteristic with the wavelet transform, has discussed the image data compression characteristic with the wavelet transform and the superiority in this foundation, and has produced the wavelet image encoding process. Contrasted and researched the classics algorithm, which based on wavelet transform---EZW algorithm, the SPIHT...