一种基于M带小波变换的多幅图像去噪算法

Multiple image copies denoising algorithm based on M-band wavelet transform.

基于小波的图像去噪算法是目前图像去噪研究的一个热点。大多数的研究考虑的都是单幅图像样本的情况,在基于图像的多幅含不同均匀噪声拷贝的两带小波去噪方法的基础上,把多幅含噪图像拷贝的两带小波去噪方法推广到了M带,提出了一种基于M带小波变换的多幅图像去噪方法。实验结果表明该方法的去噪效果要优于相同条件下的两带小波的去噪效果。

Wavelet-based image denoising algorithms are of great interest in the study of image denoising.But most of the researches have focused on only one set of observations.Based on study of the algorithm proposed by S.Chang which recovers an image from it＇s multiple noisy copies by two-band wavelet transform,the authors extend it to M-band wavelet transform,and propose a multiple image copies denoising algorithm based on M-band wavelet transform.The experimental results show that the denoising performance is better than the two-band wavelet under the same condition.