This study focused on the measurement of displacement on port structures with the combination of a marker and a vision sensor installed on the fixed location. The proposed algorithm optimizes the intrinsic parameters of the vision sensor using genetic algorithm and then measures the 6-DOF displacement automatically tracking feature points on the marker extracted through the image process. Experimental studies were carried out by using a general IP camera and a marker including 12 feature points. The distance between the camera and the marker was 10 m. Several types of displacements were induced with the precise motion stage, and the results showed that the measurement errors might be reasonable less than 3.5% in rotation and 4.0% in translation. This system was applied on the lock gate located in Incheon port, which corresponds to Type 1 facilities in Korea. The same camera was fixed at 24 m from the marker and the marker was moved by controlling the precise motion stage. The errors of measurements were less than 1 mm and 0.5 degrees, which were precise and accurate enough to evaluate the displacement in the field considering the present evaluation standard for drift, settlement, and slope on port facilities.
Read full abstract