Research Of Fractal-Based Image Compression Arithmetic

In the image compression field, fractal image coding is one of the new coding methods. Fractal image coding has been paid great attention because of its novelty, high compression rate, resolution independence and rapid decoding. Fractal image coding and encoding has been developed,however its potentiality has not been excavated out. Its compression is not perfect because the compression ratio is a little low and the time of coding is long. Therefore, enhancing the coding speed of fractal image compression has important realistic significance and bright prospect.The paper discussed the fractal image compression of the basic principles, analyzed the existing fractal image compression algorithm systematically and the advantages and disadvantages of the basic fractal image compression algorithm. Then the emphasis was programming for the adaptive quadtree fractal image compression algorithm. Through setting different error threshold, it tested the encoding time, image compression ratio and the peak value signal-to-noise ratio after compressed the selection of different image.In the basic theories of fractal compression, the domain block must carry out a series of affine transformation and each range block coding must match the transformed domain block from library. This leads to the encoding time is not good to deal with real-time .To solve the question, a diamond-based search algorithm for fractal image compression coding was adopted. And the diamond search algorithm, the nearest match and the error criteria were the point.We adopted the error of matching formula to eliminate the impact of scaling factor and compensation factor. According to the diamond search algorithm for large diamond-shaped template and small diamond-shaped template, using the matching search criteria reduced the search time and improved the coding effect. It reduced the coding complex compared with the block-fixed fractal image compression. we used the some typical images to compress. And the results showed that under the regular error threshold, this algorithm has advantages over adaptive quadtree fractal image compression algorithm at improving the coding effect.