基于Tetrolet变换域核维纳滤波斑点噪声抑制

针对图像中广泛存在的斑点噪声,提出了一种基于Tetrolet变换域抑制方法。该方法首先对噪声图像进行三层Tetrolet分解,获得低频和高频分解系数;然后对于高频Tetrolet分解系数采用基于核方法的自适应维纳滤波器进行滤波处理;对于低频Tetrolet分解系数设计出一种可调节参数的自适应阈值函数进行噪声抑制;最后进行系数重构。实验结果表明,该算法具有较好的滤波效果。

Speckle noise suppression algorithm of kernel Wiener filtering based on Tetrolet transform domain

A new image noise suppression algorithm is put forward so as to filtering the speckle noise.Firstly,the image noise is conducted for three layer Tetrolet transform,high-frequency coefficient and low-frequency coefficient is obtained.Then,the kerne Wiener filter is used to filter the noise of the high-frequency coefficient;a adjustable parameter threshold value function is designed to deal with the noise in low-frequency coefficient.Finally,the filtering coefficient is reconstructed.The experiment shows that the the algorithm is superior to wavelet transform domain hard domain threshold,wavelet transform soft threshold,Tetrolet transform domain threshold value method.