Drone Photographing and Image Recognition-based Online Monitoring of Insulator Contamination in Transmission Lines
The conventional online monitoring feature extraction of insulator contamination on transmission lines adopts an area coverage method,which leads to low monitoring contrast and increased relative error of the final results.Therefore a method for online monitoring of insulator contamination in transmission lines based on drone photographing and image rec-ognition was proposed,and corresponding design and practice were carried out.According to the current measurement re-quirements,image acquisition is carried out first,using a multi-point approach to improve the contrast of monitoring and achieve multi-point extraction of pollution features.On this basis,an image recognition model for online monitoring of transmission line insulator contaminations is designed,and dynamic tracking calibration is adopted to achieve online moni-toring.The test results show that for the selected 5 transmission line insulator contamination monitoring,compared with the groups of conventional methods such as HHT-based and 4G-based monitoring,the designed method ultimately ob-tained a monitoring relative error of less than 2.5 .This indicates that with the assistance of unmanned aerial vehicle image recognition technology,the designed online monitoring method is more effective,targeted,and has controllable application errors,which to some extent improves the accuracy and stability of monitoring.
万方数据
维普
