In the autonomous inspection of the power system,unmanned aerial vehicles(UAVs)need to be able to capture high-quality images under various environmental conditions in order to accurately detect and identify potential problems.In order to achieve this goal,the improved Faster Region-Convolutional Neural Networks Faster(Faster R-CNN)algorithm is adopted,with different shooting backgrounds,lighting conditions and shooting distances taken into account,which are the typical environmental conditions affecting the UAV's shooting effect.In order to improve the recognition accuracy,the images taken by the UAV are normalized to ensure that the image quality meets the requirements of the recognition algorithm.The improved algorithm can quickly and accurately detect the poles and towers,insulators,fittings and other typical power equipment components in the image.An adaptive shooting algorithm is developed,which can automatically adjust the shooting parameters of the UAV according to the influence of environmental factors such as wind speed and light,so as to ensure that the captured images are centered and clearly visible.Through the above methods,the high-definition adaptive focus shooting of the key parts such as the pole and tower,the wire,the insulator and the fittings is realized,and the quality of the inspection image is greatly improved.This method,which combines advanced algorithms and adaptive control,significantly improves the autonomous inspection capability of unmanned aerial vehicles in complex environments,ensuring the safe and stable operation of the power system.
关键词
电力系统/自主巡检/无人机/改进的Faster/R-CNN算法/电力系统/杆塔/绝缘子/金具
Key words
power system/autonomous inspection/unmanned aerial vehicle/improved Faster R-CNN algorithm/power system/pole and tower/insulator/fitting
维普
万方数据