一种新型高斯噪声滤波算法

为了更有效滤除数字图像中的高斯噪声,提出了一种新型滤波算法.该算法首先将含有高斯噪声的图像进行二维小波分解,得到高频和低频小波分解系数;然后保留低频小波系数不变,对高频小波系数通过维纳滤波器进行滤波,并进行小波系数重构;最后将重构图像进行多尺度小波分解,通过设定新的阈值和判别函数,弱化不重要的小波分解系数,并进行小波分...

A New Gaussian Noise Filtering Algorithm

In order to filter the Gaussian noise of digital image more effectively,a new filtering algorithm was proposed.Firstly,the image which contains Gaussian noise was decomposed two-dimensional wavelet,obtaining high frequency and low frequency wavelet coefficient.Secondly,retain the low-frequency wavelet coefficient unchanged,at the same time the high-frequency wavelet coefficient was conducted wiener filtering,the wavelet coefficients were reconstructed.Thirdly,the reconstruction image was implemented multi-scale wavelet decomposition,setting new threshold discriminant function so as to weaken unimportant decomposition coefficients.Finally,wavelet decomposition coefficients were reconstructed.By using this filtering algorithm,wiener filtering,wavelet threshold value method and average filtering on Gaussian noise denoising processing.The experiments show that PSNR value of the filtering algorithm is much higher than the other three methods.