摘要
双目立体视觉模型通过模拟人眼可以实现对目标距离的测量。为了获得水上船舶实时的运动状态,提出了一种基于双目立体视觉的船舶轨迹跟踪方法。首先,通过摄像机标定、线性空间点三维重建可以测得以相机为中心到船舶的距离,得到船舶的部分运动轨迹;其次,在双目立体视觉测距系统的基础上采用常速(Constant Velocity,CV)模型的方法对船舶运动建模;最后,对建立的船舶运动模型利用强跟踪卡尔曼滤波(Strong Tracking Kalman Filter,STKF)船舶轨迹跟踪的方法跟踪船舶的轨迹并估算目标船舶实时的运动状态。实验结果表明,基于双目立体视觉的船舶轨迹跟踪的方法能有效地跟踪船舶轨迹且适用于工程应用的需求。
Binocular vision model can achieve the target distance measurement by simulating the human eye. In order to obtain real-time motion state of sea vessel, this paper proposed a ship trajectory tracking method based on binocular vi- sion. Firstly, the method of the camera calibration and the linear space points in 3D reconstruction can measure the dis- tance between the camera and the ship when taking the camera as the center,and calculate part of the trajectory of the ship. Secondly, this paper adopted the method of constant velocity（CV） model to establish the ship motion model. Final- ly, the strong tracking kaiman filtering（STKF） method of ship trajectory tracking is used to track ship trajectory and estimate the motion state of target ship in real time for the established ship motion model. Experiments show that the ship trajectory tracking method based on binocular vision can track the ship trajectory effectively and is suitable for en- gineering applications.
作者
黄椰
黄靖
肖长诗
姜文
孙毅
HUANG Ye HUANG Jing XIAO Chang-shi JIANG Wen SUN Yi(School of Computer Science and Technology, Wuhan University of Technology,Wuhan 430063, China School of Navigation,Wuhan University of Technology,Wuhan 430063, China)
出处
《计算机科学》
CSCD
北大核心
2017年第1期308-313,共6页
Computer Science
基金
国家自然科学基金资助项目(51579204)资助
