Structure displacement extraction method based on image feature matching

The structural displacement is an important part of the health monitoring of engineering structures. The traditional displacement identification uses wired or wireless measurement systems for data acquisition, the process of arranging sensors and acquisition equipment is tedious and time-consuming. To solve the problem, a non-contact displacement measurement method based on image matching algorithm is proposed to improve the efficiency of structural displacement recognition. Consumer cameras are used as acquisition devices, the improved ORB feature matching technology is used to track the motion of the target feature points and extract the target displacement. The proposed method is applied to the displacement measurement experiment of structural model, and the results are compared with those of traditional ORB algorithm and grating scale displacement sensor. The results of experiment show that the proposed non-contact measurement method based on the improved ORB image matching technology has high measurement accuracy, and can accurately identify the structural displacement characteristics.