摘要
为了提高图像变化检测的准确率,缩短图像变化检测的检测时间,文章提出了一种新的合成孔径雷达(SAR)图像的变化检测算法。首先对实验图像进行对数变换,把图像中的乘性噪声转化成加性噪声。然后利用非线性各项异性扩散方程(PM扩散方程)的去噪算法对变换后的图像进行去噪,在对数域利用差值法获取差异图像。考虑到差异图像统计特性复杂,难以建立准确的模型,文章利用FCM算法对差异图像进行聚类,得到变化检测结果。实验结果表明,文章提出的检测算法具有较高的准确率和较短的检测的时间。
In order to improve detection precision and shorten the detection time, a new unsupervised change de- tection in SAR images is proposed. The logarithm transform is used to transform the images into logarithm domain, while multiplicative noise in images is transformed into additive noise. The PM diffusion equation is used to reduce the additive noise of the transformed images. The difference operator is utilized to get difference image. Since statistical characteristics of the difference image is complex, it is difficult to establish a accurate model for difference image. FCM clustering is selected to cluster the difference images into two classes. The experimental results show that the pro- posed algorithm can improve the detection accuracy and shorten the detection time.
作者
王雪国
贾振红
覃锡忠
杨杰
NIKOLA Kasabov
WANG Xue-guo JIA Zhen-hong QIN Xi-zhong YANG Jie NIKOLA Kasabov(College of Information Science and Engineering, Xinjiang University, Urumqi 830046,China Institute of Image Processing and Pattern Recognition, Shanghai Jiaotong University, Shanghai 200400, China Knowledge Engineering and Discovery Research Institute, Auckland University of Technology, Auckland 1020, New Zealand)
出处
《激光杂志》
北大核心
2017年第4期28-31,共4页
Laser Journal
基金
教育部促进与美大地区科研合作与高层次人才培养项目(2014-2029)
关键词
对数变换
PM扩散方程
SAR图像
变化检测
FCM聚类算法
logarithm transform
P- M nonlinear diffusion model
SAR image
image change detection
FCM clustering
