Prediction and Estimation of Vessel Movement Parameters Based on Video Monitoring

Video monitoring as an important part of bridge health monitoring is getting more and more attention in practice. The acquirement of vessel movement parameters through video monitoring is a practical but challenging problem encountered in the engineering field. An approach combining kalman filtering with iteration algorithm was proposed to practically predict and estimate the vessel movement parameters using the monitored spatial data of vessels. The vessel state vectors are obtained through moving object tracking method based on optical flow. Numerical simulations as well as practical analysis are demonstrated to verify the accuracy and feasibility of the proposed approach. The results indicate that the proposed approach can accurately predict and effectively estimate movement parameters of vessels.