科研之友
微信
新浪微博
Facebook
分享链接
摘要
In order to improve the accuracy of synthetic aperture radar (SAR) image change detection and have good change detection results, this paper proposes a method based on non-subsampled shearlet transform (NSST) detection in SAR images, for unsupervised changes. First, two original, registered images go through smoothing filtering, and then the normalized difference ratio method is used to obtain the difference image. After that, NSST is used to decompose the distance map into low frequency and high frequency sub-bands. The low frequency sub-bands are processed using linear enhancement, and the high frequency sub-bands are processed using the adaptive threshold method. Then, inverse NSST is used to obtain the enhanced difference figure. Finally, the fuzzy local information C clustering (FLICM) algorithm is used for clustering the pixels of the image into changed and unchanged sections. The experimental results show that the algorithm can effectively improve the accuracy of remote sensing image change detection, and it is not affected by the statistical distribution of the changed and unchanged sections. The proposed algorithm has higher detection accuracy than the FLICM, DWT2-FLICM, and NSCT-FLICM algorithms.
