Eight Direction Improved Sobel Algorithm Based on Morphological Processing in 5G Smart Grid
Wu Dong, Dili Peng, Xu Liu, Tao Wang, Jiao Long · 2021
Aiming at the 5G application scenario of smart grid, it is difficult to quickly process massive videos and pictures and extract effective edge information. In this paper, based on the analysis of traditional Sobel edge detection operator, aiming at its shortcomings of noise sensitivity and limited direction template, we improve it. Firstly, morphological processing is added to smooth the image to improve the sensitivity of traditional Sobel operator to noise. Secondly, this paper proposes an algorithm of Gaussian filtering combined with improved difference formula. At the same time, Sobel operator template is improved to eight direction Sobel template based on Manhattan distance. Simulation results show that the proposed algorithm can significantly improve the noise sensitivity and edge ambiguity of existing algorithms in the aspects of boundary point suppression, anti noise, and image edge optimization.