Still-camera multiview Spectral Optical Flow Imaging for 3D operating-deflection-shape identification

Image-based displacement measurement methods have long been recognized for their many practical advantages. Their adoption is often only limited by technical aspects of the image-based displacement measurement process [1]. With advances in high-speed imaging technology and high-speed video processing techniques many of these shortcomings have been eliminated, and digital-image processing techniques have become a valid option for highly accurate operational and modal analysis [2]. Image-based methods are particularly well-suited to cases in which a high spatial resolution is required, and the use of To measure high-frequency 3D vibrations, multi-camera, high-speed imaging hardware is normally required. An alternative using still-frame cameras was recently introduced with the Spectral Optical Flow Imaging (SOFI) method. In this research, the SOFI method is extended to multiview measurements of spatial operating deflection shapes. This is achieved by utilizing harmonically controlled illumination to perform an analogue Fourier transform on image-intensity data in multiple camera views. The obtained multi-view displacement spectra are combined with geometrical data to perform frequency-domain triangulation and reconstruct spatial deflection shapes. By introducing additional camera views into the image-based measurement, its field of view is extended and the signal-to-noise ratio of the final result is increased. For linear, time-invariant mechanical structures under stationary excitation, full-field 3D measurements of high-frequency vibrations can be performed using a single still-frame monochrome camera. The proposed method identifies displacements in the frequency domain directly on the camera sensor, resulting in orders-of-magnitude smaller data sizes and post-processing times compared with conventional multiview image-based methods. (c) 2020 Elsevier Ltd. All rights reserved.

Automated summary

Image-based displacement measurement methods have been widely recognized for their practical advantages, and advances in high-speed imaging technology and video processing have helped overcome many shortcomings. Multi-camera measurements can improve signal-to-noise ratio, while identifying displacements in the frequency domain directly on the camera sensor can significantly reduce data sizes and post-processing times.