An Image Contour Extraction Model Based on Depth Information and Phase Information

Edge extraction in images is very important for the understanding of image semantics. The Fourier components at various step, line, roof and other edge feature points in an image have the largest homophily, and the underlying features of the image are extracted using the phase congruency information of the image, which is consistent with the human visual mechanism. Meanwhile, different targets are at different depths in the image. Integrating the edge features at different depths is easier to obtain meaningful target contours. In this paper, we propose a contour extraction model based on depth information and phase information, and the accuracy and superiority of the algorithm in this paper are proved by natural image experiments.