摘要
由于图像二进小波变换在每次分解时不进行下抽样,所以其表示同小波级数相比是冗余的,且图像二进小波变换的部分系数扰动不会带来重构图像的严重失真。因此,在相同的误判概率下,基于二进小波变换的图像去噪效果会好于基于小波级数变换的图像去噪效果。基于这个思想,该文提出了DWID方法,将基于小波级数的图像去噪方法推广到基于二进小波变换的图像去噪,比较了DWID同基于小波级数去噪效果。实验表明,DWID比小波级数去噪效果有明显改善。
Since downsampling does not take place in image dyadic wavelet transform at each level,image representa-tion in dyadic wavelet domain compared with wavelet series reconstruction is very redundant and part of disturbance of image dyadic wavelet coefficients in transform domain will not lead to serious distortion.Therefore,with the same error decision probability,the better reconstruction can be expected.Based on this idea,this paper extends the existing wavelet-based Image denoising approaches to the dyadic wavelet-based image denoising(DWID).Numerical experiments show that DWID can significantly improve the Power Signal-to-Noise Ratio.
出处
《计算机工程与应用》
CSCD
北大核心
2003年第8期9-12,共4页
Computer Engineering and Applications
基金
国家自然科学基金资助项目(编号:69975015)
关键词
图像去噪
二进小波变换
门限估计
Image denoising,Dyadicwavelet-based image denoising,Threshold estimation