摘要
为减少遥感图像中的散斑噪声,获取较为准确、完整的变化信息,提出一种基于ROF模型半隐式去噪的SAR图像变化检测算法。对于两时相含噪图像进行对数变换,利用Rudin等提出的ROF模型,采用半隐式差分格式对含噪图像进行去噪;针对去噪后的两时相图像进行差值运算得到差异图;利用模糊C均值聚类算法对差异图聚类,得到变化检测结果图。实验结果表明,该算法具有较高的检测精度和时间运行效率。
To reduce the speckle noise in remote sensing images and obtain more accurate and complete information of changes,an algorithm of SAR image change detection based on semi-implicit denoising of ROF model was proposed.The two-phase noisy images were transformed logarithmically.The ROF model proposed by Rudin et al was used to denoise the noisy images with semi-implicit difference scheme.The denoised two-difference operation was used to get the difference image.The fuzzy C-means clustering algorithm was used to cluster the difference image to get the change detection result map.Experimental results show that the proposed algorithm has high detection accuracy and time running efficiency.
作者
娄雪梅
贾振红
杨杰
Nikola Kasabov
LOU Xue-mei;JIA Zhen-hong;YANG Jie;Nikola Kasabov(College of Information Science and Engineering,Xinjiang University,Urumqi 830046,China;Institute of Image Processing and Pattern Recognition,Shanghai Jiao Tong University,Shanghai 200240,China;Knowledge Engineering and Discovery Research Institute,Auckland University of Technology,Auckland 1020,New Zealand)
出处
《计算机工程与设计》
北大核心
2019年第4期1052-1057,共6页
Computer Engineering and Design
基金
教育部促进与美大地区科研合作与高层次人才培养基金项目(2014-2029
2016-2196)