Image Compression with a Hybrid Wavelet-Fractal Coder

  • Jin Li ,
  • C.-C. Jay Kuo

IEEE Transactions on Image Processing |

Published by IEEE

A hybrid wavelet-fractal coder (WFC) for image compression is proposed in this research. The WFC uses the fractal contractive mapping to predict the wavelet coefficients of the higher resolution from those of the lower resolution and then encode the prediction residue with a bitplane wavelet coder. The fractal prediction is adaptively applied only to regions where the rate saving offered by fractal prediction justifies its overhead. A rate-distortion criterion is derived to evaluate the fractal rate saving and used to select the optimal fractal parameter set for WFC. The superior performance of WFC is demonstrated with extensive experimental results.