Structural Health Monitoring of Buildings Using Computer Vision: A State-of-the-Art Review

Structural Health Monitoring (SHM) is essential for building safety, durability, and functionality. Buildings, as key components of the built environment, suffer from cracking, spalling, corrosion, and moisture damage. Traditional SHM approaches like vibration-based measurements, non-destructive testing (NDT), and manual inspections are reliable. However, they are expensive, slow, and difficult to use at scale. Recent developments in computer vision (CV), powered by advances in machine learning (ML) and deep learning (DL), have enabled modern, automated, and contactless inspection systems capable of detecting structural defects with high precision. This paper reviews the state of the art in computer vision applications for SHM of buildings. It focuses on the evolution of image processing, ML and DL architectures, and new 3D and multimodal systems. The paper categorizes common building defects, lists datasets for algorithm training and validation, and gives examples from recent studies. Finally, the review identifies current obstacles and suggests future research directions. It focuses on integration with drones, Building Information Modelling (BIM), the Internet of Things (IoT), and Digital Twin technologies.