A Vision-Based System for Structural Displacement Measurement

Current structural displacement measurement methods for structural health monitoring (SHM) are based on displacement data of acceleration, strain, laser doppler vibrometer, Light Detection and Ranging (LiDAR), total station, and Global Navigation Satellite System (GNSS) measurements.However, these methods are time consuming, labor intensive, limited in spatial and temporal resolution, costly and restricted to certain applications.For these reasons, a new method to measure structural displacements is needed.This study examines a novel structural displacement measurement method using a vision-based system coupled with computer vision algorithms.To test and evaluate the performance of the proposed method, seven tests were performed with varying focal lengths and 89 distance measurements using a calibrated meter stick.Results show that the error in a distance measurement decreases to within 0.02% as the measured distance increases for a fixed focal length.Furthermore, the error in a distance measurement decreases to within 1.15% as the focal length increases.Therefore, the proposed methodology is recommended for efficiently measuring structural displacements ranging from 1 mm to 1000 mm with errors less than 1.15%.