A Study Of Color Image Compression Algorithm Based On Integer Wavelet Transform And Color Protecting

As a major branch of image processing, image compression gets abroad application in many fields. With the development of modern science and technology, the compressing ratio isn't the only object for the comparison of compressing algorithm performance, the optical result of reconstructed image is thought more and more of, especially for some fields which request higher color reproduction, protecting image color information with might and main to get the highest ratio and the best compression performance is requested.In this thesis, a new algorithm with the function of color protection based on color image important coefficient scanning firstly is proposed. The strategy of color information protection contains three parts: the first is the choice of image color space. An appropriate color space under the principle that the image energy is centralized on light component and placed well-proportioned on two chroma ones in the new space is selected; second is wavelet transform. Beginning with image data wavelet analysis, Some key technologies such as the impression that wavelet base makes on image coding, the principles of choosing wavelet base and image edges processing methods are emphasized. Wavelet lifting is analyzed at length, lastly the integer wavelet base used in the algorithm is given and the method of symmetry periodical extension is used for the image edges; last is wavelet coefficient encoding. Considering optical, mentality and wavelet coefficient characteristic, the relationships among the components and color error minimized controlling, based on EZW algorithm, multi transforms are used and combined, some new improvement such as a new important coefficient scanning firstly mechanism, zero-tree extension are added. Then the coder is designed.In the performance analysis of the algorithm, image quality is measured objectively using PSNR (peak signal-to-noise ratio), MCE (mean color error), PAE (peak absolute error)and so on. Image optical comparison is given. Through the comparison with JPEG and JPEG2000, the results display the good compression and the color protecting performance, up to the scheduled aim of the thesis.