3D 점군데이터 기반 철도 터널 단면 파라미터 자동 추출 기술 개발 = Development of an Automated Extraction for Railway Tunnel Cross-Section Parameters Based on 3D Point Cloud Data

다국어 초록 (Multilingual Abstract)

Due to the limited accessibility and low illumination inside railway tunnels, conventional visual inspection methods face significant limitations, as they rely heavily on the operator’s subjective judgment and often fail to produce consistent results. Recently, LiDAR (Light Detection and Ranging)-based spatial measurement technology has been actively introduced in the field of infrastructure maintenance to overcome these challenges. In this study, a method is proposed to automatically extract tunnel cross-sections from LiDAR-acquired 3D point cloud data and to compute key geometric parameters for each section. The proposed approach was validated using 3D point cloud data from the Hwanghak Tunnel on the Gyeongbu High-Speed Railway line, yielding an average RMSE of 3.427 mm in rail height estimation and 2.17 mm in cross-sectional surface analysis. The results demonstrate that the proposed method can serve as a fundamental technology for tunnel cross-section analysis and maintenance automation, with potential applications in the development of intelligent, digital twin-based railway infrastructure management systems.