基于Mean-Shift算法的道路车辆跟踪  

Road Vehicle Tracking Based on Mean-Shift Algorithm

作　　者：吴宣康 徐祥鹏 

机构地区：[1]南京莱斯信息技术股份有限公司,南京210008

出　　处：《指挥信息系统与技术》2016年第3期86-90,共5页Command Information System and Technology

摘　　要：基于视频检测和跟踪技术的道路车辆信息采集是智能交通新兴研究方向。采用了基于均值偏移(Mean-Shift)算法的视频车辆跟踪方案,实现道路车辆的精确检测和稳定跟踪。采用Kalman滤波预测跟踪位置,以提高跟踪鲁棒性。最后,通过试验结果表明该方案的有效性和实时性。The road vehicle information acquisition based on video detection and tracking technology is an emerging research direction of the intelligent transportation systems（ITS）.A video vehicle tracking scheme based on Mean-Shift algorithm is adopted.The accurate detection and the stable tracking on the road vehicle are realized.A Kalman filter method is introduced to predict tracking position,thus enhancing the tracking robustness.Finally,experimental results show the effectiveness and the real-time capability of the scheme.

关 键 词：车辆检测 视频跟踪 均值偏移算法 KALMAN滤波 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]