Fractal Image Compression Based on Number of Hopping and Variance of Continuing Positive and Negative Pixels

In the present paper a fast and efficient fractal image encoding method based on sorting the contracted domain pool is presented. Two parameters are used to sort domain pool: the number of hopping and the variance of continuing positive and negative pixels in the given block. With the help of the parameters, the number of domain blocks examined by a range block is reduced dramatically, and thus, the sort of domain blocks results in a considerable acceleration of the encoding process. To compare with the method done by C.K.Lee that sorts the contracted domain pool in terms of their local variances, the proposed method is proved to be a better way to improve the quality of pictures with a little more time in encoding.