Fractal Images Compressing by Estimating the Closest Neighborhood with Using of Schema Theory

Abstract

Problem statement: One of the methods used for compressing images and especially natural images is by benefiting from fractal features of images. Natural images have properties like Self-Similarity that can be used in image compressing. The basic approach in compressing methods is based on the fractal features and searching the best replacement block for the original image. Approach: In this research with this attitude that the best blocks are the neighborhood blocks, we tried to find the best neighbor blocks; this search process was improved by using genetic algorithms and Schema theory. Compressing images can be considered from three approaches, first the speed of compressing, second: quality of image after Decompressing and the third: Compressing rate. In this research in addition to reducing time for compressing, the desired quality and rate of compressing were also obtained. Results: Totally genetic algorithm increase the speed of convergence for reaching the best block, but using this human knowledge that neighbor blocks always have the best chance to be replaced, were included in genetic algorithms first through neighborhood and then schema theory and this significantly decrease the time for producing a compressed images. Conclusion: Using this algorithms show the improvement in fractal compressing images comparing to other technique in compress ratio, time complexity and quality of final image parameters.