A METHODOLOGY OF BRIDGE INSPECTION INFORMATION MANAGEMENT THROUGH OBJECT RECOGNITION AND NAMING OF 3D MODEL
Junwon Park1, Sang Il Park2, Bong-Geun Kim3 and Sang-Ho Lee4
1) Ph.D. Candidate, Department of Civil & Environmental Engineering, YONSEI University, Seoul, Korea.
2) Ph.D. Candidate, Department of Civil & Environmental Engineering, YONSEI University, Seoul, Korea
.
3) Ph.D., Division of R&D Center, TAESUNG S&I, Seoul, Korea.
4) Dr. Eng., Prof., Department of Civil & Environmental Engineering, YONSEI University, Seoul, Korea
Abstract: In this study, we present an algorithm that conducts unique naming for each object through the object recognition focused on 3D straight bridge model. For numerical recognition of a three-dimensional object the algorithm utilizes centroid point, solid object on the local coordinate system, it classifies the object feature set by classifying the object and member and also recognizes bridge direction. Through a unique name containing information of span, member types, and the sequence of objects was given. Also, the suitability of naming algorithm which uses object feature set applied to bridge model like truss bridge model with different coordinate systems in examined. Also, naming based-on object feature set was carried out for the real three-dimensional bridge model and was applied to the module on local server, and mobile device. Through the comparison of the developed naming method based on object identification and the conventional method based on field inspection, it was shown that the conventional field inspection work can be effectively improved.
Keywords: 3D bridge model, Solid object recognition, Object feature set, Naming algorithm.
Bibliographical Reference:
Junwon Park, Sang Il Park, Bong-Geun Kim and Sang-Ho Lee. “A METHODOLOGY OF BRIDGE INSPECTION INFORMATION MANAGEMENT THROUGH OBJECT RECOGNITION AND NAMING OF 3D MODEL.” In Proceedings of International Conference on Civil and Building Engineering Informatics (ICCBEI 2015), 99. Tokyo, Japan, 2015.