摘要
提出了一种适用于城市道路车道线检测及车道偏离报警的方法。首先,设定道路图像感兴趣区域,对感兴趣区域进行图像预处理,提出了一种基于先验知识的改进Hough变换来提取出车道标识线参数;其次,改进了车道线跟踪策略,采用Kalman滤波器建立感兴趣区域,用最小二乘法对感兴趣区域内的车道线进行拟合,得到车道线最优预测值;最后,提出一种无需进行摄相机参数标定的车道偏离报警模型,该方法计算简单,报警精确度高。实验结果表明:在城市道路上,车道检测和偏离报警平均准确率可以达到93%以上,平均处理速度42 ms/帧左右,具有实时性和较强的鲁棒性,能够满足城市道路车道偏离报警的要求。
This paper presents a robust method for urban road detection and departure warning. Firstly, in order to reduce the computation, this paper sets the interested regions and extracts their essential edge information, and then an improved Hough transformation based on prior knowledge is used for processing the lane markings parameters. Secondly, this paper adopts Kalman filter to predict the interested regions and uses least squares method to fit the lane lines in these regions, which can improve the robustness of lane detection. Finally, this paper proposes a simple and accurate lane departure warning model which doesn＇ t need the camera parameter to warn the lane departure. The experimental results in urban road show that this proposed method can obtain a 93% average accuracy in lane detec- tion and departure warning, and produce a frame in 42 ms, which can meet real-time and robustness requirements.
出处
《电子测量与仪器学报》
CSCD
北大核心
2015年第5期685-691,共7页
Journal of Electronic Measurement and Instrumentation
基金
国家自然科学基金(61175075)
国家863(2012AA112312)
交通运输部科技(201231849A70)项目
