Multi-layer heterogeneous multi-modal convolutional neural network integrated robot inspection method

• Main Authors: YIN LINFEI, LIU DONGDUAN, MO MINGSHAN, LU QUAN, GAO FANG, BU XIANGPENG
• Format: Patent
• Language: Chinese; English
• Published: 07.09.2021
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

According to the multi-layer heterogeneous multi-mode convolutional neural network integrated robot inspection method, an intelligent inspection robot is applied to inspection of a power system, the problems caused by manual inspection can be solved, and the automation level of the power system is improved. According to the method, an image processing technology and deep learning are introduced into intelligent recognition of power system equipment, field image acquisition is carried out through an inspection robot, acquired multi-layer heterogeneous information is subjected to a three-layer multi-mode convolutional neural network graph, so that the problem of automatic data analysis and recognition under massive field working conditions is solved, and the problems of switch, pressing plate, and the like are solved. Indicator lamp screen cabinet information is detected and recognized, the time-consuming and labor-consuming problem in an electric power inspection task is solved, and the detection precision of