Algorithmic study demonstrates automated tree hazard identification in power transmission corridors, indicating high-accuracy aerial spatial monitoring.
Key Points
To develop an efficient and automated method for identifying dangerous tree encroachments near high-voltage power transmission lines using aerial 3D point cloud data.
Captured transmission corridor spatial data using drone-mounted LiDAR and performed data denoising and feature enhancement via principal component analysis (PCA).
Constructed an integrated 3D R-tree and octree indexing structure to rapidly locate regions within a predefined minimum safe distance buffer.
Applied Euclidean clustering with cylinder k-point constraints to extract hazardous tree crowns and fitted the transmission corridor lines using RANSAC-based least squares to calculate clearance distances.
The integrated spatial indexing and clustering framework successfully separated hazardous vegetation point clouds from surrounding corridor terrain.
Experimental verification demonstrated that the proposed spatial fitting and distance calculation method delivers efficient, accurate, and sensitive automated detection of corridor vegetation hazards.