SAR image edge detection using curvelet transform and Duda operator

Edge detection plays a critical role in image reconstruction of remote imaging systems. To gain better quality detection, a novel method is proposed that combines the curvelet transform and Duda operator in a special fashion for processing synthetic aperture radar (SAR) images. To extract the coarse regions in the image, the curvelet transform is first applied to the image and then, after some processing, the Duda operator locates the edges precisely inside the regions. The superiority of the proposed method is demonstrated for Pi-SAR image data.