为降低基于机器视觉车道偏离预警系统的误警率，提出一种考虑驾驶人换道意图的车道偏离预警系统。运用SteerableFilter 方法对所采集的道路图像信息进行滤波，运用局部搜索区域法提取车道线参数，运用基于图像信息的识别方法检测车辆的车速、转向信号、车道偏离状态以及驾驶人的头部动作状态，判断驾驶人的换道意图，建立了车道偏离预警的决策算法及系统。应用Matlab 软件对实车采集得到的视频进行算法验证和系统仿真试验，结果表明，提出的车道偏离预警决策算法是可行的，该预警系统将有意识与无意识的车道偏离区分开，从而能有效屏蔽在驾驶人有意识偏离车道时的误报警，具有更高的可靠性。
In order to reduce the false alarm rate of the lane departure warning system based on machine vision, a new lane departure warning system considering the driver's intention of changing lanes is put forward. In the proposed system, the collected road image information is filtered with Steerable Filter method, the road parameters are extracted using the local search algorithm, and the speed of motor vehicle, turning signal, lane departure situation as well as the driver's head movement are detected to judge the driver's lane departure intention. A lane departure warning decision algorithm is develped. Verification of the algorithm and simulation experiment are done using Matlab, with videos collected from vehicles. The results show that the proposed lane departure warning system is feasible. The system is more reliable, which can distinguish an intentional lane departure from unintentional lane departure and effectively avoid false alarm when the driver intends to change lanes.
[1] 刘志强, 蔡策, 童小田. 我国道路交通安全现状分析[J]. 公路交通科 技, 2001, 18 (2): 70-73. Liu Zhiqiang, Cai Ce, Tong Xiaotian. Currently situation analysis of road safety in China[J]. Journal of Highway and Transportation Research and Development, 2001, 18(2): 70-73.
[2] Liu C, Subramanian R. Factors related to fatal single-vehicle run-offroad crashes[R]. Washington: National Highway Traffic Safety Administration, 2009.
[3] Fardi B, Scheunert U, Cramer H, et al. A new approach for lane departure identification[C]//Intelligent Vehicles Symposium, 2003 Proceedings IEEE. New York: IEEE, 2003: 100-105.
[4] Chen M, Jochem T, Pomerleau D. AURORA: A vision-based roadway departure warning system[C]//Proceedings of Human Robot Interaction and Cooperative Robots，1995 IEEE/RSJ International Conference on. New York: IEEE, 1995: 243-248.
[5] Dagan E, Mano O, Stein G P, et al. Forward collision warning with a single camera[C]//Intelligent Vehicles Symposium, 2004 IEEE. New York: IEEE, 2004: 37-42.
[6] 王荣本, 余天洪, 郭烈, 等. 基于机器视觉的车道偏离警告系统研究综 述[J]. 汽车工程, 2005, 27(4): 463-466. Wang Rongben, Yu Tianhong, Guo Lie, et al. A survey on the research of vision-based lane departure warning system[J]. Automotive Engineering, 2005, 27(4): 463-466.
[7] Kim Z W. Robust lane detection and tracking in challenging scenarios[J]. Intelligent Transportation Systems, IEEE Transactions on, 2008, 9 (1): 16-26.
[8] Borkar A, Hayes M, Smith M T. A novel lane detection system with efficient ground truth generation[J]. Intelligent Transportation Systems, IEEE Transactions on, 2012, 13(1): 365-374.
[9] Doshi A, Trivedi M M. On the roles of eye gaze and head dynamics in predicting driver's intent to change lanes[J]. Intelligent Transportation Systems, IEEE Transactions on, 2009, 10(3): 453-462.
[10] 刘志强, 於以辰, 汪澎. 驾驶员注视行为模式识别技术研究[J]. 中国 安全科学学报, 2013, 6(23): 80-85. Liu Zhiqiang, Yu Yichen, Wang Peng. Study on driver's gaze behavior pattern recognition technology[J]. China Safety Science Journal, 2013, 6(23): 80-85.
[11] Hefenbrock D, Oberg J, Thanh N, et al. Accelerating Viola-Jones face detection to FPGA-level using GPUs[C]//Field-Programmable Custom Computing Machines, IEEE Annual International Symposium on. New York: IEEE, 2010: 11-18.
[12] 段鸿, 程义民, 王以孝, 等. 基于Kanade-Lucas-Tomasi算法的人脸特 征点跟踪方法[J]. 计算机辅助设计与图形学学报, 2004, 16(3): 279-283. Duan Hong, Cheng Yimin, Wang Yixiao, et al. Tracking facial feature points using Kanade-Lucas-Tomasi approach[J]. Journal of Computer-Aided Design & Computer Graphics, 2004, 16(3): 279-283.
