一种基于优化参数的非局部均值滤波算法

摘要

针对非局部均值滤波算法中难以找到一个全局最优的滤波参数h的问题,给出一种新的该参数的优化方法,并将其应用于传统非局部均值滤波算法的改进.首先基于SUSAN算法提取噪声图像的边缘信息,然后在大量实验的基础上,利用线性回归和非线性回归分析方法建立h与边缘信息、噪声方差之间的优化模型.最后,将基于该优化模型的非局部均值算法应用于多幅图像的去噪处理中.实验结果表明,新算法改善了传统非局部均值算法的去噪性能,取得了良好的滤波效果.%Aiming at the difficulty of finding a global optimum filtering parameter h on non-local mean filtering algorithm, a novel method of optimisation of parameter h is presented and is applied to the improvement of non-local mean algorithm. First, the edge information of noise image is extracted based on SUAN algorithm. Then, on the basis of lots of experiments, the optimum model of h in relation with edge information and noise variance is set up by linear regression and nonlinear regression analysis methods. Finally, several noise images are processed by non-local mean algorithm based on this optimum parameter model. The experimental results show that the proposed algorithm improves the denoising performance of traditional non-local mean algorithm and a good filtering effect is obtained.