ABSTRACT
Operational modal analysis (OMA) usually involves conventional sensors (wired or wireless) for vibration measurement of structures to determine the modal characteristics. This type of measurement requires sensors installation which can be time consuming and labor intensive. Video camera-based noncontact vision sensing has emerged as a promising alternative to conventional contact sensors for structural dynamic response measurement and health monitoring. Such vision sensors are relatively easy and agile to set up and provide significantly more high-spatial-density measurements where any pixel could become a measurement point. Recently, advanced computer vision algorithms like phase-based video motion magnification technique have been successfully developed to effectively process large-scale image/video data for extracting detailed structural dynamic information under small amplitude motion. The objective of this paper is to explore the potential use of phase-based video motion magnification method for OMA of civil structures. At first, a preliminary study on is carried out on a two-degree-of-freedom frame model at laboratory. Using image processing technique, the corresponding displacement is extracted from the magnified video frames and the natural frequency of the structural system is computed and verified with the reference wireless accelerometer sensors. Later, the method was implemented to practical field measurements using for modal identification and vibration measurements of existing light poles on a highway bridge. Results from both experiments suggest that the method could be used to obtain natural frequency of the structures under ambient condition where vibration amplitudes are very small and invisible.
