一种有效的去除图像混合噪声的算法

An Efficient Algorithm for Mixed Noise Removal in Image

提出一种有效的去除图像混合噪声的算法.该方法包括空间域的脉冲噪声去除和小波域的高斯噪声去除两个阶段.空间域的脉冲噪声去除利用一种加权平均的同组滤波算法进行,完成图像初始滤波;小波域的高斯噪声去除则利用NeighShrink阈值化方法对小波系数进行收缩,其中,为了提高峰值信噪比和增强视觉效果,修正了NeighShrink方法中小波系数的收缩因子.最后,对所提算法进行了仿真研究,仿真结果表明所提算法能有效去除图像中的脉冲和高斯混合噪声,并较好地保存了图像细节.

An efficient algorithm for mixed noise removal in image is studied in this paper including space impulse noise removal and wavelet Gaussian noise removal.Firstly,a weighted algorithm based on peer group filter is given to filter impulse noise.Secondly,the NeighShrink method is used to shrink the wavelet coefficients.In order to improve the visual quality of the denoising image,the shrinkage factor for the wavelet coefficient is modified by introducing an extra parameter.At last,simulations are conducted on the presented algorithm,and the simulation result shows that the presented algorithm can not only remove mixed Gaussian and impulse noise in image efficiently,but also preserve image edge information.