Adaptive fractal image compression

This paper presents a fractal method for image compression. The method makes use of self-similarities existing in an image at different resolutions under a set of affine transformations and exploits this kind of redundancy in order to achieve compression. We propose a fast search method to improve the encoding time and an adaptive block splitting strategy which is more flexible than the quadtree methods found in literature. We also show how the method can be used for colour image compression by coding the luminance component with high accuracy and the chrominance components with lower accuracy, leading to better compression factors.