Damage Information Mapping from Images to BIM Models for the Detection of Ancient Stone Bridges
Year: 2025
Type: article
Abstract: . Ancient stone bridges, as precious cultural heritage, carrying the deep historical memory and craft traditions of the Chinese nation, are susceptible to spalling, cracks and other damages due to the long-term impact of traffic load, environmental erosion, natural aging, and human damage. The traditional manual detection method is inefficient, subjective and lacks a standardized assessment system. Therefore, this paper proposes an intelligent detection and digital management method of ancient stone bridge damage by integrating image recognition and Building Information Modeling (BIM), and constructs an integrated technical process of “recognition-conversion-mapping”. The method uses deep learning model to realize pixel-level segmentation and feature extraction of bridge damage, develops multi-scale spatial coordinate conversion method to map the real damage location into BIM model, and establishes relevant damage components to realize visual management.
Authors: Yingkui Sun, Miaole Hou, Huiwen Chen, Censhan Gao, Aliea Nallbani
Language: English
Subfield: Civil and Structural Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4414812828