A CNN-based edge detection algorithm for remote sensing image

With the development and applications of satellite remote sensing technology, the edge detection accuracy of remote sensing image is increasingly high. As the gray remote sensing image has a lot of noise, even image brightness, and vague edge, a novel edge detection algorithm based on cellular neural network (CNN) is presented In the algorithm, image filtering, gray threshold segmentation, dilation and erosion, and edge detection using CNN are performed for remote sensing image successively. The experimental results show that, compared to the traditional edge detection algorithms of Sobel operator and Canny operator, the proposed edge detection algorithm can not only effectively eliminate the influence of the noise on edge detection, but also quickly detect the complete image edge.