光流法, 位移, ORB特征点, FAST特征点, LK-金字塔光流法

The optical flow method is affected by factors such as light intensity, fog, temperature, etc., making it difficult to ensure the accuracy of displacement measurement in on-site application. This article proposes an improved structural displacement monitoring method that combines ORB feature points with optical flow method. Firstly, an adaptive threshold is set in the stage of FAST feature point detection, unstable edge response points are removed using Hessian matrix, and non-maximum suppression is used to evenly distribute feature points, to improve the accuracy of feature point matching. Then, on the basis of improved ORB feature points, the LK-pyramid optical flow method is introduced to achieve high- precision tracking of feature points. Finally, the method is validated through laboratory experiments and on-site engineering applications. The results show that the method proposed in this paper can

improve the accuracy of feature point matching, meet the requirements of engineering measurement, and serve as an effective means of displacement monitoring for structures such as dams.

optical flow, displacement, ORB feature point, FAST feature point, LK-pyramid optical flow