Study On Hybrid Algorithm For Fractal Image Coding

Image compression is an important branch of data compression, and its research has made important progress. In order to achieve a more effective compression, many new ideas have been proposed, fractal image coding is one of them. Fractal image coding, for its high compression ratio, resolution independent, and the decoding process has nothing to do with the source image and other advantages, was concerned by more and more peoples. With its advantages, the fractal image compress, wavelet transform and DCT transform were considered as the next generation image compress technology.In this paper, we study the hot issues of fractal image encoding. Focusing on the problem of classic fractal image coding lack of competitiveness, compared to the current widely used encoding method, this paper discuss the fractal image coding combined with wavelet algorithm and the fractal image coding combined with the genetic algorithm, propose an improved hybrid algorithm based on genetic algorithm. This thesis is mainly divided into two parts:1. In the research of fractal image coding, we analyzed several accelerated methods of the basic fractal image coding, and proposed a new accelerated method. By using the error formula, which can derive an inequality of block's variance, this inequality is the basis to kick out the dismatched domain block, it also can reduce the computational and further improve the encoding speed.2. In the study of hybrid fractal image coding algorithm, we analyzed two kinds of hybrid algorithm of wavelet transform and genetic algorithm combined with fractal image coding. Based on hybrid algorithm of the genetic algorithm combine with fractal image coding, we proposed an improved genetic algorithm of enhanced optimal solution search ability and used its search ability to replace the complicated block matching process of fractal image coding, and combined with the contrast of contract scaling factor this method can limit the search space.