Some observations from the first application of a computer vision monitoring of displacements in multi-story buildings during shake-table tests

This paper presents an effective computer vision methodology for monitoring displacements in multi-story buildings during shake-table testing, as a convenient alternative to conventional contact sensors. The proposed system, in its simplest configuration, uses only two video cameras: one installed on the roof pointing downward to simultaneously measure absolute and relative horizontal displacements of targets at different floor levels, and one at ground level as a stationary reference to acquire the roof displacement by tracking a target near the internal camera. This setup provides i) redundant roof displacement measurements and ii) compensation for possible noise affecting the internal camera during shaking. The methodology was first applied during the shake-table testing of a full-scale, six-story mass timber building on the 6-DOF outdoor shake-table (LHPOST6) at the University of California San Diego. Based on the results, a critical discussion of the methodology’s performance is presented.