计算机技术与发展2016，Vol.26Issue(6)：16-19,4.DOI:10.3969/j.issn.1673-629X.2016.06.004
一种改进权重的非局部均值图像去噪方法
A Non-local Means Denoising Algorithm with Improved Weighted Function
摘要
Abstract
The NLM denoising uses self-similarity of image between neighborhood to construct weight,thus to achieve the effect of image restoration. The non-local means denoising model is introduced in this paper,especially for the exponential function which is the kernel function in the original non-local means denoising algorithm. And through the analysis of several new weighted kernel function,integrat-ed the advantages and disadvantages of them,a new weighted kernel function is put forward. Then research on the bilateral filtering algo-rithm,reference of its advantages,and combined with new previous kernel function,an improved weighted function is obtained,proposing a new formula of weight,getting an improved non-local means denoising algorithm. The proposed method has been evaluated on testing images with various levels noise. Numerical results show that compared with the traditional non-local means algorithm, the improved method can protect the edges,highlight the geometry features and texture,make the denoising image become more clear and result in a better effect. The proposed method improves the denoising performance as well as the preservation of structure information.关键词
图像去噪/非局部均值去噪/加权核函数/高斯噪声Key words
image denoising/non-local means denoising/weighted kernel function/Gaussian noise分类
信息技术与安全科学引用本文复制引用
黄玲俐..一种改进权重的非局部均值图像去噪方法[J].计算机技术与发展,2016,26(6):16-19,4.基金项目
四川省青年基金(2011JQ0003) （2011JQ0003）
