Abstract
We present a method for distance measuring planar displacements and rotations with image processing methods. The method is based on tracking the intersection of two non-parallel straight segments extracted from a scene. This kind of target can be easily identified in civil structures or in industrial elements or machines. Therefore, our method is suitable for measuring the displacement in some parts of structures and therefore for determining their stress state. We have evaluated the accuracy of our proposal through a computational simulation and validated the method through two lab experiments. We obtained a theoretical mean subpixel accuracy of 0.03 px for the position and 0.02 degrees for the orientation, whereas the practical accuracies were 0.1 px and 0.04 degrees, respectively. One presented lab application deals with the tracking of an object attached to a rotation stage motor in order to characterize the dynamic of the stage, and another application is addressed to the noncontact assessment of the bending and torsional process of a steel beam subjected to load. The method is simple, easy to implement, and widely applicable.
Highlights
In the last years, computer vision systems have been demonstrated to be an efficient tool to accurately track movements and vibrations, mainly in those applications where contact measurement is inadequate
The majority of applications rely on displacement tracking, which is a key parameter in structures in order to evaluate their security under loading
Because of its linear formulation, it has a strong dependence on the scene luminance. It has a limited spatial accuracy, and, it only stands for displacement tracking but not for object rotations
Summary
Computer vision systems have been demonstrated to be an efficient tool to accurately track movements and vibrations, mainly in those applications where contact measurement is inadequate. Local displacements can be described through cross-correlation, which allows an investigator to obtain the relative position between two identical images [1]. This correlation can be applied on specific targets as well as on distinct object parts. The advantages of cross-correlation are widely known. Because of its linear formulation, it has a strong dependence on the scene luminance. It has a limited spatial accuracy, and, it only stands for displacement tracking but not for object rotations. Solving the third problem is not so easy, and it is necessary to usually make use of a mathematical transformation of both the scene and the target prior to the cross-correlation operation [4,5,6]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
