Rapid integration strategy for oblique photogrammetry terrain and highway BIM models in large-scale scenarios

• Published in: Automation in construction Vol. 177; p. 106354
• Main Authors: Li, Jiaheng, Han, Feng, Shi, Long, Liu, Zelong, Wang, Chengxiang, Fan, Yikun
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.09.2025
• Subjects: Automated processing; Boolean operation algorithm; Highway building information modeling; Oblique photography; Scan line fill algorithm; Terrain integration; Terrain integration; Oblique photography; Highway building information modeling; Boolean operation algorithm; Scan line fill algorithm; Automated processing
• ISSN: 0926-5805; 1872-7891
• DOI: 10.1016/j.autcon.2025.106354

The integration of Oblique Photogrammetric Digital Models (OPDMs) with Highway Building Information Modeling (HwyBIM) models to construct Three-Dimensional (3D) scenes is vital in the digital twin technology used for linear infrastructure, especially in geological hazard simulation and monitoring. However, owing to large data volumes and complex geometric features, no efficient method currently exists for rapid OPDM integration processing. This paper proposes a fast extraction method for Donflicting Tile Models (CTMs) based on the Scan Line Filling (SLF) algorithm to reduce redundant data loading during integration. An improved Surface-Volume Boolean Operation Algorithm (F-V BOA) is developed to process individual Tile Models (TMs), along with a 3D Model Boundary (3D-MB)-driven automatic TM integration plugin. The proposed approach is validated through a real-world highway case study and compared with conventional approaches, demonstrating its effectiveness and efficiency. The proposed approach can significantly contributes to OPDM integration in large-scale linear infrastructure projects. Future work will incorporate CPU/GPU parallel computing for further optimization. •This study addressed the challenges of large data volume and time-consuming processes in OPDM integration for large-scale scenarios.•A CTM rapid localization and extraction method was proposed, which effectively reduced redundant data loading by 58.8 %.•Single TM integration was implemented based on an F-V BOA, and a 3D-MB-driven automatic TM integration plugin was developed.•The proposed approach demonstrated significant improvements in both integration quality and computational efficiency.