Edge detection of SAR images using incorporate shift-invariant DWT and binarization method
Can Wang, Weimin Su, Hong Gu, Hua Shao · 2012
In this paper, a fast unsupervised noise-robustness edge detection method in the synthetic aperture radar (SAR) images is presented. Within this method the whole edge detection process is divided into two simple tasks. Firstly, the input image is decomposed with shift-invariant discrete wavelet transform (DWT) and edge enhancement is done by the information of the subbands of the SAR images. Secondly, a fast unsupervised hybrid method of the binarization and edge detection is used for SAR images edge detection. This method incorporates speckle reduction and edge detection as a single process so that complex operations are avoided. Compared to the denoising first and then edge detection method, simulations show the proposed method achieves better performance and has low computations.