Deep Learning-Based Computer Vision for Health Monitoring in Civil Engineering

Abstract:

Health monitoring in the field of civil engineering is of great significance to ensure the long-term and stable service of infrastructure. Compared with traditional monitoring methods， the computer vision technology based on deep learning has the advantages of high efficiency and accuracy. This paper provides a systematic review on the application of the deep learning-based computer vision technology in the field of civil engineering life cycle health monitoring. First， a scientific econometric analysis of the literature in this field is conducted with the help of literature visualization software. Then， the development process of computer vision technology is briefly described， and the methods of data acquisition， data processing， and data annotation in the process of constructing deep learning data sets are summarized. Afterwards， the development and practical engineering application value of the computer vision technology based on deep learning in safety management of construction site， local damage detection of in-service structures and overall damage assessment of structures after disaster are reviewed. Finally， the future application directions are prospected.