Flower Pollination Technique Forfractal Image Compression

Abstract

Metaheuristic Optimization techniques are used to overcome the problems inherent in iterative simulations. These methodshave certain remarkable characteristics and abilities, which is that it requires only less computation time and memory to find better solutionsi.e., near optimal solutions without using any composite derivatives. Over the years, varieties of meta-heuristic optimization algorithms have been introduced and used to compress the images with minimal loss of information.The objective of image compression is to reduce the redundancy of the image and to store or transmit data in an efficient form. One such type of image compression is Fractal Image Compression (FIC). These FIC techniques commonly use the optimization techniques to find the optimal best solution.The aim of the FIC is to divide the image into pieces or sections and then finds self-similar ones. It produces high compression ratio, fast decompression in short amount of time. But the major disadvantage of Fractal Image Compression is that, it requires high computational cost and retrieves image with poor qualities.In this paper, Flower Pollination Based Optimization approach is used for fractal image compression. This optimization technique effectively reduces the encoding time while retaining the quality of the image.Here, Flower pollination algorithm (FPA) is compared with genetic algorithm(GA)and their performances are analysed in terms of compression ratio, encoding time and PSNR(Peak Signal-to Noise Ratio) value.