A homography transformation-based video stabilisation method for structural health monitoring using an unmanned aerial vehicle

摘要

Using a computer vision technique to perform vibration-based structural health monitoring has the advantages of being low cost and non-contact and thus it attracts a lot of attention and has made great progress recently. It is difficult, however, for static cameras to be placed in effective locations in many scenarios, such as the monitoring of large-scale wind turbines, bridges, etc. The use of unmanned aerial vehicles (UAVs) is a good option for dealing with this issue. Over the past ten years, the application of UAVs for structural health monitoring has been increasing dramatically. However, the nonstationary motion of a hovering UAV provides a great challenge in terms of the estimation of dynamic characteristics. Thus it is important to stabilise the video to achieve the level at which vibration can be monitored. This paper proposes a video stabilisation method based on homography transformation for the purpose of applying the use of UAVs to vision-based vibration monitoring. Firstly, a motion model is built concerning the relationship between the motion of the camera, called egomotion, and its displacement on film. Secondly, an experiment is conducted to define the characteristics of the egomotion of the hovering UAV. Thirdly, a stabilisation algorithm is developed for compensation of the egomotion.