Fast fractal image coding approach based on fuzzy classification
Jin-Shu Han · Computer-aided engineering · 2006
With the problem of expensive time spending in the fractal image encoding, an approach is proposed that 6 parameters of edge image blocks are selected as the features of image blocks. The original image blocks are classified by a fuzzy pattern recognition technology. Then, the cluster analysis technology based on part grey means is used to shrink the search ranges for optimal matching block. The simulation results show that the approach can speed up encoding time greatly with no apparent deterioration on the quality of the reconstructed image.