Fractal Image Coding by Multi-Scale Selection Based on Block Complexity

Since the conception of automatic fractal image compression, the research on this topic has grown rapidly. This work is intended to provide a new vision on this automatic process by introducing the idea of multi-scale domainpool classification based on the complexity of the image to be compressed. A preprocessing analysis of this image identifies the complexity of each image block computing its local fractal dimension. The performance of this proposition, evaluated by means of fidelity versus encoding time and amount of compression, is compared with two well-known image compression methods.