摘要
利用小波系数的相似性,提出一种基于人眼视觉特性的小波域分形混合图像编码方法。该方法是将图像进行小波分解,按照人眼对于图像边缘区、平滑区、纹理区敏感程度的不同,给这三种不同区域内图像信息所对应的小波系数赋予不同视觉权值,并分别进行分形编码。实验结果表明在保证优先编码视觉上最重要系数的同时,能够进一步提高图像复原质量。
Taking use of the similarity of wavelet coefficient in the wavelet domain,a method of wavelet and fractal image coding based on human visual system is proposed.This method seeks to make wavelet transform to the image. According to the different sensitivity of human visual system on edge,smooths region and texture of the image,gives different perceptual weights to the corresponding wavelet coefficient,and encodes the coefficient separately by fractal coding.The experiment results demonstrate that the method can improve the recovery quality of the image by encoding the major coefficient at priority.
出处
《计算机工程与应用》
CSCD
北大核心
2005年第29期52-54,共3页
Computer Engineering and Applications
基金
黑龙江省自然科学基金项目(编号:F2004-01)
关键词
小波变换
分形图像编码
迭代函数系统
人眼视觉特性
wavelet transform,fractal image coding,Iterated Function System,Human Visual System(HVS)