Eulerian fast motion identification algorithm for deformation measurement of cable-stayed bridge

Video magnification has received increased attention in structural deformation measurement. Current video magnification methods typically chose the Eulerian perspective rather than the Lagrangian, which requires tedious work for tedious parameter adjustment. However, the low computational efficiency is a prominent problem of the Euler-based method due to the complex time domain filtering processing. To address this issue, a novelty motion magnification algorithm, called Euler fast motion detection (EFMD), is proposed in this paper, which can realize real-time measurement with low computation cost. Laboratory verification and field experiments were carried out to verify the accuracy of the algorithm in deck and cable dynamic response measurement. In laboratory experiments, the error of natural frequency measured by the proposed method is within 2%. In the field experiments, the small displacements (about 1.5 mm) can be achieved by the proposed method at a distance greater than 20 m, and the error of the cable vibration is less than 4%.

Automated summary

This paper proposes a novel motion magnification algorithm called Euler fast motion detection (EFMD), which can achieve real-time measurement with low computation cost. Laboratory verification and field experiments demonstrate its high accuracy in measuring dynamic response.