基于ICA和FCM的多时相SAR图像变化检测

针对多时相合成孔径雷达(Synthetic Aperture Radar,SAR)图像提出了一种非监督的变化检测方法,首先利用独立成分分析(Independent Component Analysis,ICA)抑制相干斑噪声和减少混合像元,提高图像的分类性能.然后对独立分量图像构造差值图像进行非监督的变化检测,最后采用加入空间邻域信息的模糊C-均值聚类(Fuzzy C-Means)方法把差异图像分为变化类和非变化类,以此克服需要选择统计模型的影响.实验结果证实了该算法能有效抑制了变化检测中的虚警.

Change Detection on Multi-temporal SAR Images Based on ICA and FCM

In this paper,a novel unsupervised change detection scheme is presented to find changes in two registered Synthetic Aperture Radar(SAR) images acquired at different times.Firstly,a method based on Independent Component Analysis(ICA) is proposed to reduce speckle and mixed pixels in SAR images.It can improve the performance of image classification.Then,differencing image are extracted through ICA of SAR images to carry out unsupervised change detection.In order to overcome the limitations imposed by the need of selecting a statistical model for differencing image and considering the spatial-contextual information in the differencing image,the fuzzy clustering algorithm is used in to cluster the two groups of pixels belonging to the "change" and "no change" classes(binary mapping).Experimental results show that the proposed approach can suppress false alarms of change-detection process.