Civil Structure Monitoring and Vibration Control Based on Computer Vision

Nowadays, with the increasing number of super tall buildings and major engineering projects, their safety and stability issues are becoming increasingly prominent. This paper intends to use computer vision methods to establish a method for online monitoring of building structures and conduct real-time monitoring. This project plans to use high-precision cameras to capture subtle deformations and cracks on structural surfaces, and combine machine learning algorithms to process them, achieving real-time evaluation of structural integrity and safety. In addition, the system also integrates a set of vibration reduction devices, utilizing active control technology to effectively reduce adverse effects caused by external factors such as wind loads and traffic in the external environment. This paper combines computer vision and vibration engineering to improve the processing speed and accuracy of monitoring data, providing engineers with a better understanding and control of the mechanical behavior of structures, and ensuring public safety. The identification accuracy of mid frequency vibration modes is slightly lower than that of low-frequency modes, but overall it remains above 95%. This paper not only achieves real-time monitoring of the overall state of the structure, but also can detect hidden dangers in the early stage, greatly improving the preventive maintenance and emergency response capabilities of the equipment.