Automated extraction of tunnel electricity transmission system: An object-level approach with mobile laser scanning data
Shida Wang, Hangbin Wu, Han Yue, Lianbi Yao, Chun Liu, Haili Sun
Abstract
One of the essential works of the tunnel maintenance department is to inspect and maintain the electricity transmission system (ETS). Tunnel inspection with a mobile laser system (MLS) can automate the traditionally manual surveys. In this study, we proposed a step-wise method for automatically extracting ETS in MLS data. First, in the approximate extraction stage, we used edge-based and fitting-based segmentation algorithms to remove the ground and lining, respectively. Then, in the precise extraction stage, power transmission lines were accurately extracted at the object-level using a proposed spherical-stepping-cluster algorithm, and the supporting fixtures were extracted using density information and connection characteristics. The proposed method was validated through experiments on a batch of point cloud markings with ground-truth data and comparisons with existing methods. The average F-score for the datasets was 94.2 %, indicating that this research provides a new paradigm for extracting tunnel ETS.