Recognition and Locating of Damaged Poles in Distribution Network Through Images Shot by Unmanned Aerial Vehicle (UA V)

The disaster recognition and locating are important for the rush repairment after disaster happened. And there have been some advanced facilities applied for the image collecting, such as unmanned aerial vehicle (UAV). To accelerate the image processing, the intelligent recognition and automatic location solution are both considered instead of the manual way. Firstly, the model based on yolov5 is construed to realize the recognition of damaged poles in distribution network, marked by a rectangle box. Then, the coordinate transformation is adopted to derive the positions of damaged poles in real world with the inputs of coordinates in images and the necessary information extracted from images. With the locations of damaged poles collected, the fast disaster locating can be realized. At last, the practical images are chosen to testify the performance of intelligent recognition and location solution.