Intelligent Tilt Monitoring Method for Pylon Using High Precision Terrestrial Laser Scanning

The precision and stability of pylon tilt state monitoring using unmanned aerial vehicle LiDAR data are unsatisfactory. Terrestrial laser scanning technology (TLS) with high-density point cloud is introduced to solve this problem. After the preprocessing, including filtering, extraction, and registration, the TLS pylon point cloud is divided and projected to two directions, which are vertical and horizontal to transmission lines. An improved YOLO v8 network is applied to extract key points for tilt calculation from projection images, and the tilt state is computed based on the geometric relationship of the key points. The experimental results show that the proposed method differs from the reference value by only 0.05%<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</inf>, and increases the results with better stability compared to current algorithms for UAV LiDAR data, which indicates that it can be applied to the tilt state monitoring missions of key pylons.