Research And Application Of Fractal In Image Compression

The purpose of image compression coding is to represent images with few bits, maintain the quality of decoding images. The decoded images can apply to certain application situations. Fractal image compression is a new image coding method which developed in recent years, and has been studied extensively.Fractal theory applied to image processing that creates a fractal image compression technology. Fractal image compression coding has achieved a wealth of application results. It use the image itself has a certain degree of self-similarity, which can be expressed by the iterative function parameters. Further by generating a function of parameters that have been encoded, thus achieve the purpose of image compression, fractal image compression method of the outstanding advantage of the compression and decoding faster than higher. Exist in real life, a large number of images do not have a good self-similarity, this classic method is to divide the image into sub-blocks, and separately calculate the fractal image compression coding. This approach first rule of the shape of the image is divided into sub-block sets, and the non-overlapping sub-block, eg. 8×8image block. According to sub-blocks self-similarity, after a series of rotation and symmetry transform. Encoding this sub-block iteration function parameters will receive the fractal image compression codes.The existing algorithm is the segmentation of different sizes, respectively to be sub-block and the parent block, since each block affine transformation. The process of conducting the search matches will cost a lot of time that limit the application of fractal image compression. In this paper, fractal coding improved quality and reduced the compression time. The method which based on calculate the fractal dimension of images divide the image to sub-blocks. The encoding of fractal image method will improve the compression quality and time consume.First of all, parallel computing fractal dimension of the image than the existing method have a larger rate of increase. Secondly, the use of fractal dimension as an index to divide the image to sub-block, can receive high similarity sub-blocks. Finally, the proposed parallel computing based on fractal dimension algorithm improving the picture quality while also improving the coding rate.