Research On Image Compression Based On Contourlet Transform And Fractal

With the development of society, the communication course is turning fast.Some techniques, such as digital TV, internet multimedia communication, conference TV, visual telephone, remote sensing image transmission, image data base, data storage for fingerprint identification system and so on, are used widely. Meanwhile, the volume of image data is so huge that it is difficult to implement image transmission and images storage in limited bandwidth and storage capability. Therefore, people are hunting after constantly new techniques and methods for image compression. Wavelet and Contourlet transform have character of multiresolution analysis (MAS) and fit to the properties of visual system.Wavelet, Contourlet and Fractal technique are imported into image compression fields, which promote the new development for image compression. This thesis contributes to research the still image compression methods based on Contourlet transform and fratal, and mainly includes the following aspects.Firstly, we review the development of image compression techniques. The basic theory and thought of Wavelet and Contourlet transform are introduced.The embedded image compression method based on Contourlet transform and the coding method of morphological representation of non-redundant Contourlet transform data are introduced.Secondly, a coding method of adaptive threshold fractal and zerotree based on non-redundant Contourlet transform is introduced. The experimental results show that with the same compression ratio, the proposed algorithm can get better quality of image than the Wavelet image compression algorithm based on fractal coding and zerotree.Thirdly, to overcome the limitation of the vector quantization coding algorithms based on Wavelet in image edge preservation, a novel vector quantization method based on non-redundant Contourlet transform is presented.Experimental results show that the edges of the compressed images obtained by our proposed algorithm are clearer compared to the vector quantization coding algorithms based on Wavelet.Finally, the Krawtchouk moments is used to fractal coding,The coding method of adaptive threshold fractal based on Krawtchouk moments and the method of adaptive classification fast fractal based on Krawtchouk moments are proposed. Experimental results are given and compared with the results of the other fractal image compression methods.