基于改进权重非局部均值图像滤波方法与实现

Research on the Non-local Mean Image Filtering Algorithm Based on Improved Weight

图像滤波是计算机图像处理领域中极为重要的预处理环节,目的是消除混杂在图像中的干扰因素,改善图像质量,强化图像表现特征。在传统非局部均值滤波算法基础上,提出了基于改进权重的非局部均值图像滤波算法,以欧式距离高斯加权为基础,配以图像之间的自相似性,在图像领域灰度值的矩阵间使用,充分地将图形领域间的自相似性发挥出来。实验结果表明,基于改进权重的非局部均值图像去噪算法比传统的非局部均值去噪算法保持更有效的图像结构信息。

Image filtering is an important step in the field of computer image processing, the purpose is to eliminate the confounding factors in the image, improve the image quality, and strengthen the image performance. The non local mean denoising uses the self similarity between the image domains to construct the weights, and then to restore the image, the smaller is the distance, the greater is the weight. This paper proposed the non lo- cal means denoising algorithm based on improved image weights, the Euclidean distance weighted Gauss based, with self similarity between images and its use in the field of image gray value matrix, full graphic field of self similarity between play. Experimental results showed that the non local mean image denoising algorithm based on the improved weight is more effective than the traditional non local mean denoising algorithm.