Supervised Classification of Power Lines from Airborne LiDAR Data in Urban Areas

摘要: Automatic extraction of power lines using airborne LiDAR (Light Detection and Ranging) data has been one the most important topics for electric management. However, this is very challenging over complex urban areas, where are in close proximity to buildings trees. In paper, we presented a new, semi-automated versatile framework that consists four steps: (i) line candidate point filtering, (ii) local neighborhood selection, (iii) spatial structural feature extraction, (iv) SVM classification. We introduced corridor direction filtering multi-scale slant cylindrical features extraction. detailed evaluation involving seven scales types 26 features, two datasets, demonstrated use individual 3D points significantly improved The experiments indicated precision, recall quality rate classification more than 98%, 98% 97%, respectively. Additionally, showed our approach can reduce whole processing time while achieving high accuracy.