Pyramidal wavelet decomposition provides a hierarchical data structure for image representation-wavelet subtree

which is composed of the coefficients at different resolution and different orientation but with same spatial location. In this paper

the conventional fractal coding in spatial domain is extended to wavelet domain by employing the self-similarities between different wavelet subtrees through proper affine transformation. Further

in order to achieve a good trade-off between image quality and compression ratio.an adaptive partition algorithm is given which can split the wavelet subtree into children subtrees according to local image complexity. Experimental results show that the new hybrid image compression algorithm can obtain much better coding performance in terms of PSNR and subjective quality as compared to JPEG at roughly the same compression ratio.