基于边缘检测的自适应非局部均值去噪算法

摘要

An adaptive non-local means (ANLM) algorithm for image denoising is proposed to solve the problem that the results of denoised images are over smoothed in the non-local means (NLM) method.The proposed algorithm extracts edge contents of noisy images,compares the edge structure similarity of the current block with its neighborhood,and then,determines the structure features of the image area in which the current block is located.And two thresholds are set to select filtering parameters adaptively.Meanwhile,the weight given to the pixels to be denoised is adjusted.Experimental results show that the ANLM combines the edge structure information and the gray information effectively.Comparisons with the traditional NLM algorithm show that ANLM method preserves more details.%针对图像去噪处理中非局部均值(NLM)算法造成的平滑图像细节丢失问题,提出基于边缘检测的自适应非局部均值去噪(ANLM)算法.利用canny算子提取图像边缘结构图,比较待处理像素所在图像块和邻域图像块边缘结构的相似度,判断当前图像块所处图像区域的结构特征,设定相关阈值,自适应选取滤波参数,调整像素在平均过程中的权值.实验结果表明,ANLM算法相比NLM算法,能更有效去噪,可保留更多的边缘结构特征,细节信息更为丰富.