Research on multi-level point cloud classification method of underground tunnel

Abstract

Underground tunnel engineering is complex systematic engineering. Classifying the ground objects inside the tunnel automatically and accurately is crucial in tunnel construction surveys. Aiming at this problem, this paper proposes a multi-layer underground tunnel point cloud classification method, which uses the hierarchical clustering structure to deal with the original tunnel point cloud. It extracts four specific ground objects: tunnel surface, track, platform, and pipeline inside the tunnel step by step. Experiments show that this paper’s multi-level tunnel point cloud classification method can accurately extract these four types of ground objects. The average accuracy of the projection plane in each experimental area is not less than 95.00%, and the average accuracy of point cloud classification is not less than 92.63%, which can better meet the needs of the underground tunnel internal construction survey.