Development of Skeletal Coordinate Extraction and Skeletal Estimation Technology for Drone-Mounted LiDAR Point Clouds

Abstract

For drones to fly autonomously or fly beyond visual line of sight, it is necessary to prepare coordinates for latitude, longitude, and altitude. However, current route design based on equipment drawings and measured one point at a time in manual flight has problems with accuracy and having high workload. In this research, we developed a technology to segment the aerial point cloud obtained by LiDAR into each element using deep learning, extract skeletal coordinates, and skeletal estimation.