Computer Vision-based Displacement Measurement Method With Arbitrarily Positioned Camera



Published Jul 14, 2017
Junhwa Lee Soojin Cho Sung-Han Sim


Displacement is broadly used in structural health monitoring as a meaningful indicator of structural status. Measuring structural displacement using conventional devices such as the linear variable differential transformer is challenging due to practical issues such as equipment inaccessibility, insufficient measurement accuracy, or expensive system cost. Recently, computer vision-based displacement measurement methods have been developed to build high field applicable measurement system. However, the existing computer vision-based displacement measurement
techniques still have practical issues such as absence of camera installation point, marker attachment, or lightinduced error. This paper presents a computer vision-based displacement measurement system focusing on solving camera positioning issue. The proposed method adopts homography transform to allow cameras to be positioned at arbitrary point. A laboratory scale experiment was conducted to verify accurate displacement measurement of the proposed method with the arbitrarily positioned camera.

Abstract 36 | PDF Downloads 67




Hartley, R., & Zisserman, A. (2003). Multiple View Geometry in Computer Vision. Cambridge university press.
Lee, J. J., & Shinozuka, M. (2006). A Vision-based System for Remote Sensing of Bridge Displacement. Ndt & E International, vol. 39, pp. 425-431.
Special Session Papers