Edge Detection from Remote Sensing Images Based on Canny Operator and Hough Transform

At present, the widespread edge detection algorithms, such as Roberts edge detection, Prewitt edge detection, Sobel edge detection and Marr edge detection, detect the edge using the variation of one-order or two-order directional derivative near to the edge usually. As a result, the noises can be detected in the image regarding as the edge points, and the real edge maybe missed due to noise interference. Because the remote sensing images have a disadvantage of containing noises, the better detecting results are difficult to obtain using the above edge detection algorithms. Combined with the characteristics of remote sensing images, including rich edge information and a lot of noise mixed, this paper present a edge detection algorithm for remote sensing based on the Canny operator and Hough Transform. From the experiment introduced in the paper, the proposed edge detection algorithm has some noise immunity, and can explore the edge information accurately in the remote sensing image.