Abstract:
Based on the 5G application and practice of Guizhou power grid, aiming at the massive data information collected by 5G network, how to identify the key information effici...Show MoreMetadata
Abstract:
Based on the 5G application and practice of Guizhou power grid, aiming at the massive data information collected by 5G network, how to identify the key information efficiently and quickly based on the image intelligent recognition algorithm and realize the alarm return is the core of this paper. In view of the above image edge detection problems, this paper proposes an improved multi-dimensional Canny algorithm. In each stage, wavelet threshold denoising, improved 4-direction Sobel template, angle interpolation, fusion Otsu algorithm, genetic algorithm and double low threshold algorithm, and secondary morphological processing are used. By building a simulation platform, it is verified that the multi-dimensional improved algorithm based on Canny not only inherits the advantages of the original algorithm, but also has better performance in the aspects of algorithm efficiency, mean square error, peak signal-to-noise ratio and structural similarity compared with the traditional Canny algorithm.
Date of Conference: 23-26 April 2021
Date Added to IEEE Xplore: 21 June 2021
ISBN Information: