基于小波树和四叉树的图像分形编码算法研究  

An image fractal coding approach based on wavelet transformation and quad-tree

作　　者：牟宇飞[1] 张文普[1] 彭畑[1] 阳溢[1] 

机构地区：[1]重庆通信学院,重庆400035

出　　处：《微型机与应用》2013年第11期49-52,56,共5页Microcomputer & Its Applications

基　　金：国家自然科学基金项目(61272043)

摘　　要：图像经过多级小波分解后,各级小波子带具有明显的相似性,可利用小波变换和分形编码的优势进行图像压缩编码。对最低分辨率子带进行失真较小的均匀量化编码。对高分辨率子带,将水平、垂直和对角3个方向的小波子带结合起来考虑,形成小波树结构,设定一个阈值,小波树所有系数小于等于阈值则直接置零;大于阈值的小波树采用四叉树算法进行分形编码,如果误差小于等于阈值,则记录分形编码参数,否则进行四叉树分裂。对算法进行了Matlab仿真,结果表明,在保证一定解码图像质量的情况下,运算速度有较大提高。After the image is decomposed with scales wavelet transformation, there is similarities between different wavelet subbands. It can make use of wavelet transformation and fractal to code. The lowest resolution subband apply symmetric quantization coding. For higher resolution wavelet subbands, taking the horizontal, vertical and diagonal wavelet subbands into account and makes wavelet tree structure, a threshold is defined for the higher resolution subband. If the wavelet coefficient of the wavelet tree is not more than the threshold, it is replaced by zero. And if it is more than the threshold, it carry out fractal coding with quad-tree. Experiments show that the computing speed is improved when it assures some quality of decoding image.

关 键 词：小波树 分形编码 四叉树 压缩比 

分 类 号：TN911.7[电子电信—通信与信息系统]