Abstract: Based on short-distance piecewise self-similarity of the image, this paper proposes a real coding based on Partheno genetic algorithm to solve the matching problem of fractal image coding. Aiming at the matching of Range blocks that contain edge, it employs a direction operator to pilot the individuals moving along the edge of image. Experimental result suggests that at the costs of almost unnoticeable degeneration of decoding image quality, the proposed algorithm can achieve the speed-up of 60 times as the baseline fractal algorithm with full search.

Key words: fractal, image coding, partheno genetic algorithm, direction operator

摘要： 基于图像的近距自相似性，用一种采用实数编码的单亲遗传算法完成图像分形压缩中的最优码本匹配，针对含有边缘的Range块的匹配，通过一个导向算子使边缘R块的最优匹配搜索沿着图像的边缘进行。实验结果表明，相对于基本分形压缩算法，在基本保证解码图像主观质量的前提下，该方法的速度提高了约60倍。

关键词: 分形, 图像编码, 单亲遗传算法, 导向算子

CLC Number: 

• TP391