Image Edge Detection Based on Sobel with Morphology

Edge detection is one of the most fundamental problems in image processing and the most basic task in computer vision. For the problem that Sobel image is not clear for image edge extraction, is affected by noise, and cannot extract some image edges effectively, an improved algorithm based on Sobel image edge extraction is proposed. Firstly, the color image is converted to grayscale image to facilitate edge extraction, and the Sobel operator is used to extract the horizontal and vertical gradients of the image, secondly, the morphological transformation of the image is used, and the cross kernel is selected to do morphological processing on the extracted horizontal and vertical gradients. Finally, the experiments are conducted on the dataset, and the results show that the improved algorithm is able to extract clearer contour edges.