利用小波域平滑和边缘保留的图像去噪算法

提出了一种基于小波域低频信号平滑及高频边缘保留的图像去噪方法.由于图像噪声主要集中在小波高频子图部分,且系数较小,常用的阈值去噪法存在阈值选取过大以致连同丢失掉边缘细节信息的缺陷,因此可以对高频子图提取边缘并保留和进行软阈值去噪处理后再融合；低频子图进行自适应维纳平滑滤波,进而得到重构出的去噪图像.实验结果表明,该方法在有效去噪的同时较好的保留了原有的边缘细节信息,效果明显.

Image denoising based on wavelet smooth and edge retaining

An optimized image denoising algorithm based on wavelet smooth and edge reservation is proposed.As the image noise is mainly concentrated in the high-frequency and the coefficient is small,if the threshold is selected too large edge details will get lost.A method in which the high-frequency edge is firstly extracted and then soft threshold is used to wipe off the noises,then the low-frequency edge is smoothed by self-adaptive wiener filtering,and finaly the denoised image is obtained after wavelet reconstruction.Experimental results show that this method is effective in denoising,while it retains the original edge details.