摘要
针对图像处理过程中去噪与保边缘这一矛盾,提出了一种分类迭代保边缘的图像去噪算法。该算法充分考虑到不同密度的脉冲噪声产生的噪点分布特性,将噪声分为连续性噪声和非连续性噪声,采用大小不同的2~3个模板对原始图像迭代运算,进行噪声点恢复。利用该算法对添加有不同密度噪声的图像进行处理,结果表明,该算法在有效恢复噪声点的同时,能够很好地保留图像边缘细节信息,解决了传统算法存在滤波和保边缘的矛盾冲突,与传统滤波算法比较,该算法的峰值信噪比及去噪后图像与原图像的整体相似程度均优于传统算法。
In view of the contradiction between denosing and edge preserving in image processing,an image denoising algorithm of classification iteration with edge preserving is proposed in this paper.It takes full account of the noise distribution characteristics of impulse noise with different density and noise is divided into continuous noise and discontinuous noise.Then 2~3 templates with different sizes are used to iterate the original image to recover the noise points.The algorithm is used to process the image with different density noise.The results show that the algorithm can effectively restore the noise points while preserving the details of the image edge information,and the results show that the proposed algorithm can not only restore the noise points but also preserve the details of the edge of the image.The conflict between filtering and edge preserving in traditional algorithms is solved.Compared with the conventional filtering algorithm,the peak signal-to-noise ratio of the algorithm and the overall similarity of the denoised image to the original image are superior to the traditional algorithm.
作者
雷继海
LEI Jihai(Longdong College Electrical Engineering,Qingyang Gansu 745000,China)
出处
《自动化与仪器仪表》
2020年第3期29-32,共4页
Automation & Instrumentation
基金
国网新源浙江宁海抽水蓄能有限公司、批次计划(No.PC2017780000005)的支持。
关键词
图像去噪
噪声脉冲
分类迭代
边缘细节
image denoising
impulse noise
classification iteration
edge details