Abstract: Aiming at the problem of the over-smoothness and blurs the details, which are caused by exponential kernel function used in original non-local means algorithm, this paper proposes a cosine Gaussian kernel function based on exponential kernel function and combined with a cosine coefficient and Gaussian kernel. It is used in the weight-computing of the improved algorithm. Experimental results show the algorithm has a superior denoising performance than the original one, especially with detail information in the image, and PSNR can be improved by 1.6 dB at most.

Key words: image processing, image denoising, non-local means, weighted average, Gaussian noise, weighted kernel function

摘要： 传统非局部均值滤波算法中使用指数型加权核函数，容易导致图像细节因过度平滑而变得模糊。为此，在指数型加权核函数的基础上，采用余弦系数加权的高斯核函数，设计一种改进的非局部均值图像去噪算法，并将其应用于加权系数计算中。实验结果表明，该算法的去噪性能优于传统算法，且能更好地保留原图像的细节信息，峰值信噪比最大可以提升1.6 dB。

关键词: 图像处理, 图像去噪, 非局部均值, 加权平均, 高斯噪声, 加权核函数

CLC Number: 

• TP391