Dynamic parameters identification of 3D sandwich wall panels from phase-based video measurement via smartphones camera

Abstract

This study, unlike prior studies, utilizes a smartphone camera as a high-speed camera for output-only operational modal analysis by using video measurement and advanced computer vision algorithms. Besides, the accelerometer sensors of the three smartphones collect the acceleration data. In contrast to recent research, this study has been conducted outdoors without any additional light setup and solid color background to contrast the motions. First, multi-scale image processing and blind source separation techniques are applied to each video frame to obtain the modal coordinates. The logarithm decrement technique and Fourier transform extract the damping ratios and frequencies of dominant modes of structures. To validate the smartphone camera measurements, two 3D sandwich walls were used and their results compare with accelerometer data. After that, the modal properties of a cantilever steel beam were identified by only smartphone camera measurements. The obtained modal parameters from smartphones camera and accelerometers are in excellent agreement, results indicated that. This study showed that smartphones camera has great potential to use as a non-contact modal analysis facility.