Abstract
In recent years, asset management has been conducted for structures such as roads and bridges, being promoted by Ministry of Land, Infrastructure, Transport and Tourism (MLIT). In asset management, it is necessary to make a maintenance plan with the highest cost effectiveness by quickly grasping corrosion or flaked coating of the structures as well as their long-term changes such as distortion and transformation. However, it takes much expense in time and effort to grasp the long-term changes because it is done by inspectors' visual inspection. This research first generates a panorama image and panorama range image from the image group and range image group obtained using the range image sensor in order to detect long-term changes in a large-scale structure. Second, the structures in the panorama image and panorama range image at current and past are aligned. Then, a system is developed to detect long-term changes in color- and three-dimensional information on the structures at the same time, by comparing color- and three-dimensional pieces of information on the structures at current and past. Finally demonstration experiments are performed to verify its usability.
