# UAV Vision-Based Method for Multi-Point Displacement Measurement of Bridges

## TL;DR

A new UAV-based system accurately measures bridge displacements with high precision, enabling better safety assessments.

## Contribution

A UAV vision-based system with a computing terminal and targets achieves sub-millimeter accuracy in bridge displacement measurement.

## Key findings

- The system achieved RMSE < 0.3 mm in measuring bridge displacements under traffic loads.
- Results matched those from a Scheimpflug camera, validating the system's accuracy.
- The method enables high-precision, non-contact monitoring in complex environments.

## Abstract

What are the main findings?
A UAV vision-based multi-point displacement measurement system (integrating a UAV-mounted camera, computing terminal, and targets) is proposed to address accuracy limitations arising from UAV motion interference and camera performance constraints.Field tests on Lunzhou Highway Bridge (Guangdong Province) successfully captured full-span vertical multi-point dynamic displacements under traffic loads, with a root mean square error (RMSE) < 0.3 mm—consistent with results from a Scheimpflug camera.

A UAV vision-based multi-point displacement measurement system (integrating a UAV-mounted camera, computing terminal, and targets) is proposed to address accuracy limitations arising from UAV motion interference and camera performance constraints.

Field tests on Lunzhou Highway Bridge (Guangdong Province) successfully captured full-span vertical multi-point dynamic displacements under traffic loads, with a root mean square error (RMSE) < 0.3 mm—consistent with results from a Scheimpflug camera.

What are the implications of the main findings?
The system’s flexible deployment in complex environments enhances the applicability of high-precision, non-contact technologies for bridge displacement monitoring.It provides critical data for understanding bridge deformation behavior, supports reliable safety assessments, and advances UAV vision applications in bridge health monitoring.

The system’s flexible deployment in complex environments enhances the applicability of high-precision, non-contact technologies for bridge displacement monitoring.

It provides critical data for understanding bridge deformation behavior, supports reliable safety assessments, and advances UAV vision applications in bridge health monitoring.

The challenge of insufficient monitoring accuracy in vision-based multi-point displacement measurement of bridges using Unmanned Aerial Vehicles (UAVs) stems from camera motion interference and the limitations in camera performance. Existing methods for UAV motion correction often fall short of achieving the high precision necessary for effective bridge monitoring, and there is a deficiency of high-performance cameras that can function as adaptive sensors. To address these challenges, this paper proposes a UAV vision-based method for multi-point displacement measurement of bridges and introduces a monitoring system that includes a UAV-mounted camera, a computing terminal, and targets. The proposed technique was applied to monitor the dynamic displacements of the Lunzhou Highway Bridge in Qingyuan City, Guangdong Province, China. The research reveals the deformation behavior of the bridge under vehicle traffic loads. Field test results show that the system can accurately measure vertical multi-point displacements across the entire span of the bridge, with monitoring results closely matching those obtained from a Scheimpflug camera. With a root mean square error (RMSE) of less than 0.3 mm, the proposed method provides essential data necessary for bridge displacement monitoring and safety assessments.