Hybrid Image Compression Using Fractal-Wavelet Prediction

Based on the standard fractal transformation in the spatial domain, simple relations may be found relating coefficients in detail(high-pass) subbands in the wavelet domain. In this paper, we devise a hybrid image compression using fractal-wavelet prediction where the causal similarity among blocks of different subbands in a wavelet decomposition of the image is exploited. The proposed coding scheme consists of predicting fractal code in one subband from fractal code in lower resolution subband with the same orientation. By linear adjusting the fractal code parameters (including matched domain block position of current range block,eight isomorph type, contrast scaling α and the offset o ) in lower resolution subband, fractal code in the adjoin high resolution subband with the same orientation is approximately forecast achieved. The experimental results show that the performance of our scheme is superior for both acceptable visual decoding image quality, an average of 20 % reduction in encoding time and higher compression ratio,compared with standard Jacquin fractal coders. As a side effect, our work motivates shortened encoding time and improved bit allocation strategies for fractal coding. Key-Words: Fractal predict, Wavelet decomposition, Self-similarity, Multiresolution, Encoding