2026 Volume 7 Issue 3 Pages 564-573
This study proposes a non-contact three-dimensional measurement method for highway expansion joints to address critical issues regarding lane closure costs, traffic congestion, and worker safety. By combining time-lapse photography using a monopod-mounted camera with SfM/MVS analysis, three-dimensional point cloud data is constructed to enable desktop-based dimensional calculations of in-service lanes from a single-lane closure area. Experimental results demonstrated that the proposed method consistently satisfies the required accuracy of within 5.0 mm under recommended shooting conditions. Moreover, three-dimensional point cloud data was successfully generated even in public road environments with active traffic. A verification study involving 10 participants confirmed that the method's accuracy is highly reliable and independent of the operator's skill level. Furthermore, we demonstrated the method's potential applicability for nighttime inspections by appropriately managing illuminance and mitigating surface reflections.