摘要
针对现有滤波算法滤除图像中高斯-椒盐混合噪声后细节丢失较严重的问题,将图像块自相似性思想引入到混合噪声滤波领域,提出了基于非局部自相似性的混合噪声滤波算法。设计了图像在含混合噪声的情况下,其块相似度衡量策略及权系数的计算方式,实现了利用图像自相似性对椒盐噪声进行滤除,再结合非局部均值滤波算法处理图像中剩余的高斯噪声。实验结果表明,提出的算法可以有效地滤除图像中的混合噪声并较大程度地保留边缘、纹理等图像细节信息。
In view of the problem that the existing filtering algorithm lose more details after filtering the Gaussian and salt-pepper mixed noise in the image,the idea of image block self similarity is introduced into the field of mixed noise filtering,and a filtering algorithm for mixed noise based on nonlocal self similarity is proposed.The algorithm designs the measure strategy of the block similarity and the calculation method of weight coefficient for images with mixed noise,realizes the use of image self-similarity to filter out salt and pepper noise,and combined with the non-local means filtering algorithm to process the remaining Gaussian noise in the image.The experimental results show that the proposed algorithm can effectively filter out the mixed noise in the image and retain the image details such as edges and textures to a large extent.
作者
袁健
姜振宇
石凌峰
YUAN Jian;JIANG Zhenyu;SHILingfeng(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)
出处
《光学技术》
CAS
CSCD
北大核心
2020年第3期361-367,共7页
Optical Technique
基金
国家自然科学基金项目(61775139)。
