基于小波变换的吸引子快速图像编码  

Fast attractor image coding method based on wavelet transform

作　　者：袁占亭[1] 张秋余[1] 王晶[1] 

机构地区：[1]兰州理工大学计算机与通信学院,甘肃兰州730050

出　　处：《计算机应用》2006年第6期1357-1358,1361,共3页journal of Computer Applications

基　　金：甘肃省科技攻关项目(2GS047-A52-002-03)

摘　　要：自然图像自相似性不强,直接压缩效果不佳且压缩速度慢,但图像经过行小波变换后,不同分辨率子带图像中相同空间位置的图像块、同分辨率子带中的图像块具有相似性,而分形维数基本上反映了图像灰度变化的复杂度。在图像小波子带本身相似的基础上,提出了利用分形维数作为衡量相似图像子块是否匹配的依据,并给出了编码方法。最后通过仿真证明,该方法不仅有效的压缩了图像,而且减少了编码时间。The natural image is not strongly alike, the direct compression is ineffective and compress speed is slow, when decomposed with wavelet transform, it is divided into a number of sub-images of different frequency bands, and of different spatial orientations. The neighboring sub-images of the same orientation are strongly alike. The sub-images of different orientations, though few similarities were presented between them. Because the fractal dimension basically reflect a complications for variety of the image grey scale, so fractal dimension was used to evaluate the matching degree of similar image block. It＇s obtained an additional increase in compression ratio and reduced the code time.

关 键 词：分形图像编码 吸引子图像编码 分形雏数 图像块 相似性 小波变换 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]