Multi-view camera motion compensation in bridge displacement measurement with panoramic camera

Abstract

Ambient conditions such as wind and traffic can introduce camera motion errors in bridge displacement measurement. To compensate the motion, traditional methods use one or more cameras to track fixed reference targets, which faces limitations: single cameras have a narrow field of view, and multi-camera systems are complex, expensive, and difficult to calibrate. To address these challenges, this paper proposes a multi-view camera motion compensation method in bridge displacement measurement with panoramic camera. The main contributions include two aspects: (1) A multi-view panoramic camera is utilized to overcome field-of-view limitations and system complexity. By cube projection and independent calibration, a single panoramic camera is equivalent to a quadrature four-camera system without overlapping fields of view. (2) A motion compensation method using multi-view array targets and pose transformation matrix optimization is proposed. This method uses the refined transformation matrix and the fixed connection relationships in an equivalent four-camera system to derive a theoretical model for multi-view compensation. The model allows the use of reference targets in any direction to compensate for camera motion. The accuracy and feasibility of this method were validated through deflection measurements on a bridge scaled model and an actual bridge.