Primarily investigates the use of airborne LiDAR point cloud data during the survey and design phase of overhead transmission line projects.By training with the PointCNN framework

it extracts power lines

buildings

and other structures to create DEMs and vegetation elevation models.Based on optimized design schemes for the line

it constructs a safety sag model for power lines using conductor sag simulation and wind deflection range calculation methods.The method identifies tree obstruction hazards through three-dimensional spatial distance analysis

segments individual trees using normalized segmentation algorithms

and locally collects ground-based LiDAR point clouds or high-density airborne LiDAR power sources.It employs Yolov9 neural networks combined with RANSAC data classification to extract tree trunks and forest information.The statistical work on tree felling is advanced to the construction drawing design stage.Experimental analysis shows that this method is technically feasible and accurate

providing a basis for tree felling statistics and approval in transmission line projects.