首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于时域差动成像的车辆精细运动轨迹重构方法
引用本文:张辉,陈静萍,林军记,王浩远,龚文森.基于时域差动成像的车辆精细运动轨迹重构方法[J].汽车技术,2019(7):8-12.
作者姓名:张辉  陈静萍  林军记  王浩远  龚文森
作者单位:中山大学智能交通研究中心广东省智能交通系统重点实验室
摘    要:为实现汽车精细轨迹自感知,提出了基于时域差动成像的汽车精细轨迹重构方法。通过控制2个成像设备的差动工作,获得具有微小时间差的连续2帧路面图像,利用图像匹配技术识别和匹配其中的纹理特征点,结合标定参数、间隔时间和拍摄高度计算出汽车的速度、方向以及横摆角,最终通过速度积分法重构出汽车精细的运动轨迹。试验结果表明,所得结果精度显著优于GPS轨迹和惯性轨迹,并可有效反映车辆运动过程中的空间占用情况。

关 键 词:差动成像  汽车轨迹  速度积分  汽车运动参数  图像匹配

Method of Vehicle Detailed Trajectory Reconstruction Based on Time-Domain Differential Imaging
Zhang Hui,Chen Jingping,Lin Junji,Wang Haoyuan,Gong Wensen.Method of Vehicle Detailed Trajectory Reconstruction Based on Time-Domain Differential Imaging[J].Automobile Technology,2019(7):8-12.
Authors:Zhang Hui  Chen Jingping  Lin Junji  Wang Haoyuan  Gong Wensen
Institution:(Research Centre of Intelligent Transportation System,Guangdong Provincial Key Laboratory of Intelligent Transportation System,Sun Yat-sen University,Guangzhou 510006)
Abstract:To realize the self-perception of the vehicle detailed trajectory,this paper proposes a method of vehicle detailed trajectory reconstruction based on time-domain differential imaging.The method acquires two consecutive images of road surface with small time interval by controlling the differential operation of the two imaging devices.The image matching technology is used to identify and match the texture feature points,and the speed,direction and yaw angle of the vehicle are calculated according to the calibration parameters,interval time and shooting height.The vehicle detailed trajectory is constructed with the velocity integration method.Test result shows that the results are significantly better than the GPS trajectory and the inertial trajectory,and can effectively reflect the space occupancy during vehicle motion.
Keywords:Differential imaging  Vehicle trajectory  Velocity integration  Vehicle motion parameters  Image matching
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号