Inspection Path of UAV Power Tower Based on Multi-Sensor

This is the authors' abstract. We don't add key points for this paper.

Electric energy, as an efficient and clean energy for production, plays a very critical role in social and economic development. In order to improve the power supply capacity, under the situation of reform and opening up, my country's electric power industry has not only rapidly increased the installed capacity of power generation, but also greatly improved its power transmission capacity. In this paper, fault inspection and analysis of power towers are carried out. In transmission lines of different levels, power poles and towers often have different types and have different functions, which also makes the manual daily inspection work more complicated. Based on the multi-sensor UAV technology, this paper discusses the types of transmission towers and their functional characteristics, analyzes the types of possible faults, and proposes the locations and methods of detecting different faults. Image processing technology is used to perform various processing on pictures or video information containing gestures to achieve the purpose of recognition. It mainly refers to the steps of image segmentation, hand positioning, feature extraction and recognition. This method is a more intuitive method. The final results of the research show that with the increase of fault test points, the fault detection rate shows a trend of changing in the opposite direction. When the number of faults is 56, the detection accuracy rate is 94.56%, and when the number of faults is large, the detection accuracy there are some errors.