A camera calibration method for bridge live load identification using plane marks on the pavement

Abstract

Computer vision has become an effective way to collect the information of traffic load, which is crucial for bridge statistical analysis, safety evaluation, and maintenance strategies. To obtain the information of vehicles such as position and wheelbase, the location and orientation of the camera should be calibrated in advance, but the calibration needs measurement and sometimes is difficult to implement because of heavy traffic flow. Therefore, this paper proposes a camera calibration method using plane marks on the pavement without measurement. The plane marks are designed to be convenient and economical for construction so any bridge or road with a potential demand of vehicle monitoring can consider to prepare such marks on the pavement. The feasibility of the proposed method is verified through computer simulation and a model test.