Fractal image coding combined with subband decomposition

Abstract

In order to improve the performance of fractal image coding, this paper proposes a new coding scheme which is combined with subband image decomposition. In the proposed scheme, an input image is first decomposed to low and high resolution components by subband decomposition. The fractal block coding with adaptive range block size is performed only for the lowest resolution component. On the other hand, direct quantization and entropy coding are carried out for the other resolution components. The residual difference between reconstructed and original lowest resolution components is also quantized and entropy coded in order to enhance the coding performance. The computer simulation results show that the proposed coding scheme gives higher SNR (signal-to-noise ratio) values and better reconstructed image qualities compared to the conventional fractal block coding scheme.