多车道场景下的视频车辆测速方法  

Video-based vehicle speed estimation method for multi-lane roads

作　　者：于洋[1] 于明[1] 翟艳东[1] 王志芳[2] 

机构地区：[1]河北工业大学信息工程学院,天津300401 [2]河北科技大学理工学院,河北石家庄050035

出　　处：《河北工业大学学报》2011年第3期79-83,共5页Journal of Hebei University of Technology

基　　金：河北省教育厅重点基金项目(ZD200911);河北省科技支撑计划(11213518d)

摘　　要：针对目前大多数测速算法无法完成多车道场景车辆测速的问题,提出了一种用于多车道场景下的视频车辆测速算法.该算法对传统的虚拟线圈技术进行改进,将虚拟线圈分为分界检测区域和触发检测区域,有效的完成了多车道场景下多目标车辆的检测.同时在测速过程中,以几何中心作为特征对目标车辆进行跟踪,推导出车辆在图像中的位移.与传统方法相比,该方法实现了多车道场景下的车辆测速,有效的解决了车辆换道问题.实验证明,该算法具有良好的鲁棒性以及较好的测试效果.Aiming at the most proposed methods can not estimate the vehicle speed for multi-lane roads.A video-based vehicle speed estimation method for multi-lane roads is proposed.The proposed method improves the traditional virtual loops technology,dividing virtual loops into two regions and achieving the multi-target vehicle detection for multi-lane roads.Meanwhile,the method derives the displacement of vehicles in the image sequences by tracking the geometric center of the target vehicle in the process of vehicle speed estimation.Compared with traditional methods,this method targets the vehicle speed estimation for multi-lane roads and solves the vehicle lane-changing problem.The experiments indicate that the method is robust and effective.

关 键 词：图像处理 智能交通系统 视频测速 虚拟线圈 多车道 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]