具有边缘保持特性的零树小波图像压缩算法

基于小波变换的图像压缩算法,在含噪图像和较低码率时出现的边缘模糊现象多年来一直未能得到很好的解决.为了解决这一问题,提出一种具有边缘保持特性的零树小波图像压缩方法.首先,对图像进行小波边缘检测,确定哪些小波系数是图像的边缘特征,将其保护起来;然后,对小波变换域系数采用改进软阈值收缩方法实现去噪;最后,利用SPIHT(Set Partitioning in Hierarchical Trees)算法对图像进行压缩编码.实验结果表明,该方法不仅能获得较高的图像压缩率、较好地去除噪声,而且能在一定程度上解决边缘模糊问题,能较高地恢复图像质量.

ZERO-TREE WAVELET IMAGE COMPRESSION ALGORITHM WITH EDGE PRESERVING CHARACTER

Edge blurring phenomenon,which occurs in the wavelet-based image compression algorithms under noised image or low bit rates,remains an open question for many years.In this paper it presents a zero-tree wavelet image compression algorithm with edge preserving character to solve this problem.Firstly,the image is wavelet edge detected,the determined wavelet coefficients which are of the edge character of image will be protected,then denoise the wavelet transform domain coefficients by improved soft-threshold c...