一种基于中值-模糊技术的混合噪声滤波器amixed-电子与信息学报.pdf

第28 卷第5 期 电 子 与 信 息 学 报 Vol.28No.5 2006 年5 月 Journal of Electronics Information Technology May 2006 一种基于中值-模糊技术的混合噪声滤波器 王建勇 周晓光 廖启征 (北京邮电大学自动化学院 北京 100876) 摘 要 结合中值与模糊滤波技术，提出了一种新的图像混合噪声滤波算法。算法将受混合噪声污染的图像分为 脉冲噪声点集与含有高斯噪声的像素点集两部分，首先进行灰度极值检测，进而借助邻域纹理信息准确检测出脉 冲噪声，并以中值滤波滤除；对于含有高斯噪声的像素点则采用一种保护细节的模糊滤波器进行处理。实验结果 说明算法不仅能有效地滤除脉冲与高斯混合噪声，而且可以较好地保护图像细节。 关键词 图像处理，混合噪声，中值滤波，模糊滤波器，隶属度函数，加权均值滤波器 中图分类号：TN911.73 文献标识码：A 文章编号：1009-5896(2006)05-0901-04 A Mixed Noise Filter Based on Median-Fuzzy Technology Wang Jian-yong Zhou Xiao-guang Liao Qi-zheng (School of Automation, Beijing University of Posts and Telecommunications, Beijing 100876, China) Abstract A new algorithm of mixed noise removal for digital image is presented combing median filter and fuzzy technology. The pixels of an image are divided into two sets: impulse noise points and pixels polluted by Gaussian noise. The min-max operator is used first. Then impulse noise is detected accurately by using image texture and removed by median filter. A detail-preserving fuzzy filter is adopted to reduce Gaussian noise contained in other pixels. Experiments show that the algorithm can remove mixed impulse and Gaussian noise efficiently while protecting image details. Key words Image processing, Mixed noise, Median filter, Fuzzy filter, Membership function, Weighted average filter 1 引言 很好的效果，往往需要结合不同的方法，如文献[6]中设计了 一种结合中值与均值滤波的混合噪声滤波器；基于模糊技术 噪声滤除是图像处理的重要步骤，脉冲噪声和高斯噪声