摘要
自Ｂａｒｎｓｌｅｙ提出分形图像编码的概念，特别是Ｊａｃｑｕｉｎ给出了第一个完全由计算机自动完成的图像编码算法以来，分形图像编码得到了越来越多的关注，但分形图像压缩往往需要较长时间，这就给具体实用带来了巨大困难。本文首先给出分形图像压缩理论，然后在Ｊａｃｑｕｉｎ所提出的分形方块编码基础上提出了改进算法，包括建立小方块被匹配概率活动表，按相似度找到匹配块，隔点计算法等。针对上述改进算法，作了具体实验。实验结果表明，与基本自动分形图像编码方法相比，在保持重建图像质量的前提下，运算时间大大下降，加快了分形图像压缩速度。
Since Barnsley presented the concept of fractal image compression, especially after Jacquin proposed a fully automatic image compression algorithm, fractal image coding has attracted a great of attention. But it also needs much time, thus it is difficult to put it into effect. The theory of fractal image compression is presented,then an improved method is developed based on fractal block coding algorithm proposed by Jacquin,including setting up probability list of small block,finding matching block according to similarity and calculating point by point.This paper does experiment with the above mentioned improved algorithm. Experimental results show that compared with basic automatic fractal image coding method, this new approach needs less computation time with no obvious depression on the quality of the reconstructed image and speeds up the fractal image coding.
出处
《数据采集与处理》
CSCD
1999年第2期167-170,共4页
Journal of Data Acquisition and Processing
