摘要
针对城市交通路口的多车道车辆数量统计问题,提出基于车道映射矩阵的车道划分方案。根据车道映射矩阵分布情况判断车道中心线位置,利用相邻车道中心线间的距离对车道进行划分,与车辆计数算法结合,实现多车道车辆计数系统。该方案能够减少环境因素对车道划分的影响,在白天与夜晚情况下均取得较好效果。实验结果表明,该方案准确率为93.6%,能够适应复杂道路,多车道实时车辆计数系统准确率能够达到91%以上,在保证系统实时性的基础上,提供了较高的系统准确性和鲁棒性。
To solve the problem of multi-lane vehicle counting at urban traffic intersections,a lane division scheme based on lane mapping matrix was proposed.The position of the lane centerline was determined according to the distribution of the lane mapping matrix,and the lane was divided by the distance between the adjacent lane centerlines.The multi-lane vehicle counting system was realized by combining with the vehicle counting algorithm.The proposed scheme can reduce the influence of environmental factors,and achieves better results in both day and night situations.Experimental results show that the accuracy of the proposed scheme is 93.6%,which can adapt to complex roads.The accuracy of multi-lane real-time vehicle counting system can reach more than 91%,providing better performance on system accuracy and robustness on the basis of ensuring the real-time performance of the system.
作者
殷妍
陈庆奎
YIN Yan;CHEN Qing-kui(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)
出处
《计算机工程与设计》
北大核心
2020年第5期1481-1488,共8页
Computer Engineering and Design
基金
国家自然科学基金项目(61572325、60970012)
高等学校博士学科点专项科研博导基金项目(20113120110008)
上海重点科技攻关基金项目(14511107902、16DZ1203603)
上海市工程中心建设基金项目(GCZX14014)
上海智能家居大规模物联共性技术工程中心基金项目(GCZX14014)
上海市一流学科建设基金项目(XTKX2012)
沪江基金研究基地专项基金项目(C14001)。