Evolutionary fractal image compression using asexual reproduction optimization with guided mutation

There are many different methods for image compression which each of them satisfies a various type of purposes. Fractal Image Compression is a category of these techniques that has some specific features. This method is robust against aliasing of images in zooming, so it has multi-resolution capability. Besides, compression ratio of this method is reasonably competitive, also its decoding is fast. But the main issue of this method is the compression time which is very high because of complexity for finding self-similar blocks. So researchers have tried to mitigate computational costs with different approaches. In this paper, using an evolutionary algorithm called Asexual Reproduction Optimization (ARO) is proposed for fractal image compression. Then the main operator of this algorithm is tuned to make it more efficient versus other individual-based algorithms like Simulated Annealing (SA) and Tabu Search (TS). Finally experimental results and execution time of the proposed method, SA and full search are compared. ARO with guided mutation generates defensible outputs in very short time versus the others approaches.