计算机应用与软件2017，Vol.34Issue(7)：149-153,197,6.DOI:10.3969/j.issn.1000-386x.2017.07.028
一种改进的非局部平均图像去噪算法
AN IMPROVED NON-LOCAL MEANS IMAGE DENOISING ALGORITHM
摘要
Abstract
The pixel similarity measure of the traditional NLM image denoising algorithm is less reliable, and its key filter parameter selection and optimization value deviation is larger.Aiming at the above problems, an improved NLM image denoising algorithm is proposed.First, noise component in method noise is filtered, meanwhile preserving useful image information.Secondly, combined denoising result and the processed method noise, it is redefined the similarity function of the NLM algorithm to make better use of the original image information.Finally, the filter parameters are set by the noise standard deviation quadratic function.Compared with the traditional linear proportional mode, the selection method can obtain better parameter values under different noise intensity.The denoising results of several standard test images show that the improved NLM algorithm achieves better denoising effect, which is superior to the comparison method.关键词
图像去噪/非局部平均/方法噪声/参数优化Key words
Image denoising/ Non-local means/ Method noise/ Parameter optimization分类
信息技术与安全科学引用本文复制引用
许光宇,李玲..一种改进的非局部平均图像去噪算法[J].计算机应用与软件,2017,34(7):149-153,197,6.基金项目
国家自然科学基金项目(61471004) （61471004）
安徽理工大学博士基金项目(ZX942) （ZX942）
安徽理工大学中青年学术骨干基金项目(13335). （13335）
