Structural health monitoring based on three-dimensional point cloud technology: A systematic review

Structural Health Monitoring ensures the safety and longevity of critical infrastructure. Traditional methods face limitations such as high subjectivity and incomplete coverage. Recent advances in three-dimensional point cloud technology offer an intelligent solution due to its non-contact data acquisition, high accuracy, and full-field coverage. This paper reviews the latest research, methodologies, and applications of 3D point cloud technology in SHM, based on 223 studies published from 2010 to 2024. It outlines the entire process, from data acquisition to modeling and visualization, and compares six core methods: geometric morphology analysis, multi-temporal differential analysis, feature extraction, mapping, machine learning, and deep learning. Case studies show successful applications in damage detection, deformation monitoring, and health assessment of structures like bridges, tunnels, and historical buildings. Future efforts should optimize intelligent algorithms to shift 3D point cloud technology from ''single technological breakthrough'' to ''systemic capability development,'' supporting automated structural health monitoring and resilient urban development.