Change Detection of SAR Images Using Dyadic Wavelet Enhancement and Edge Local Information FCM

Considering the speckle noise of the SAR images which causes a large number of false-alarm and missing-alarm in the detection results, a novel method which can overcome speckle noise of SAR Images by using dyadic wavelet enhancement and local edge information FCM was proposed in this paper. First, LR different images were enhanced adaptively by dyadic wavelet to denoise and redistribute the gray levels. Then, Edge information was detected by ROEWA operator to amend weight of the pixels in the local window. It can eliminate error influence when local window slide to edge region. Finally, segmented the different images and obtained binary images. Results of simulation and real data processing demonstrate that the proposed method is effective on image detail preservation and denoising.