基于车道线虚线角点检测的行车安全视距测算模型

田顺; 田山山; 杨炜; 魏朗; 陈涛

doi:10.3963/j.jssn.1674-4861.2022.02.004

基于车道线虚线角点检测的行车安全视距测算模型

doi: 10.3963/j.jssn.1674-4861.2022.02.004

田顺^1,,
田山山²,
杨炜¹,
魏朗^1, ,,
陈涛¹

1.
长安大学汽车学院西安 710064
2.
比亚迪汽车有限公司汽车工程研究院西安 710119

基金项目:

国家自然科学基金项目 52072047

陕西省自然科学基础研究计划青年项目 2022JQ-007

中央高校基本科研业务费资助项目 300102220106

详细信息

作者简介:
田顺(1989—)，博士，讲师. 研究方向：道路交通安全. E-mail：tianshun@chd.edu.cn

通讯作者:
魏朗(1957—)，博士，教授. 研究方向：道路交通安全. E-mail：qch_1@chd.edu.cn

中图分类号: U46.1
计量
- 文章访问数: 1019
- HTML全文浏览量: 565
- PDF下载量: 74
- 被引次数: 0
出版历程
- 收稿日期: 2021-05-23
- 网络出版日期: 2022-05-18

A Model for Estimating Driving Sight Distances Based on Corner Point of Broken Line of Roadway

1.
School of Automobile, Chang'an University, Xi'an 710064, China
2.
Automotive Engineering Research Institute, BYD Automobile Limited Company, Xi'an 710119, China

摘要

摘要: 开展行车视距调查对于营运期公路安全评价至关重要，这对车载条件下行车视距检测提出了要求。针对现有基于车道线图像特征点所构建的视距模型精确度不高的问题，提出了1种以车道线虚线角点为关键特征的行车安全视距测算模型。在车载设备获取的图像预处理基础上，采用轮廓跟踪法对车道线虚线轮廓进行提取，通过设定轮廓尖锐度阈值以实现对车道线虚线角点的初步筛选；使用最大、最小距离法对候选角点进行聚类分类，将每类中尖锐度最大的点判定为真实角点；此外，结合车道线虚线图像梯形特征实现对角点的精确提取；根据成像原理的坐标转换关系，通过解构角点在世界坐标和像素坐标之间的映射关系以求解二者之间的转换矩阵，得到实际道路环境的行车安全视距测算模型；将模型所测算的行车视距与运行车速所需的行车视距进行对比，实现对不同道路线形下行车视距的评价。通过实车实验对所提行车视距测算模型进行动态和静态检测精度验证。结果表明：该模型在静态条件下的行车视距检测误差小于7%，低于采用车道线特征点提取方法检测的误差；在动态车载条件可实现行车视距的连续检测，表明在该模型能适应动态条件对行车视距的检测。该模型可实时动态检测行车视距，为营运期公路安全评价提供支撑。
- 交通安全 /
- 行车视距 /
- 虚线角点 /
- 坐标转换 /
- 图像处理
Abstract: A study of driving sight distances is critical for safety evaluation of highways, which makes it ideal for estimating driving sight distances using in-vehicle equipment. To address the low accuracy of the existing sight distance models using the feature points of lane marking images, a model for estimating driving sight distances with the dotted corner points as the important feature is proposed. Based on the images preprocessing obtained by an in-vehicle equipment, the contour tracking method is used to extract the contour of line markings, so that the initial screening of the corner points can be extracted by setting a threshold sharpness of the contour. After using the maximum and minimum distance methods to cluster and classify candidate corner points, the points with the largest sharpness in each category is determined as the"true"corner points. In addition, the accurate extraction of the diagonal points is achieved by using the trapezoidal features of the dashed line image of the lane marking. By considering the relationship between the global mapping coordinates and the pixel coordinates of the corner points, the transformation matrix between the two coordinates is settled and the estimation model of driving sight distance is developed. By comparing with estimated sight distance with the required distance at a given operation speed, the evaluation of driving sight distance of the alignment of current road segment is implemented. The dynamic and static detection accuracy of the proposed sight distance estimation model is verified by a field experiment. Study results show that the estimation errors under the static condition are less than 7%, which is lower than the traditional methods. In addition, under dynamic conditions, the errors of driving sight distance are consistent with the results of static conditions, indicating that the proposed estimation model has a good performance under dynamic conditions. Comprehensively, the model can be used to support safety evaluation of highway design and operation.
- traffic safety /
- driving sight distance /
- dotted line corner point /
- coordinate conversion /
- image processing

HTML全文

图 1 车道线虚线角点检测与提取流程图

Figure 1. Flowchart of detection and extraction of dotted line corners of roadways

下载: 全尺寸图片幻灯片

图 2 邻域判断顺序图

Figure 2. Neighboring area judgment sequence diagram

下载: 全尺寸图片幻灯片

图 3 局部轮廓线放大示意图

Figure 3. Schematic diagram of local contour line

下载: 全尺寸图片幻灯片

图 4 轮廓、角点提取局部图

Figure 4. Partial map of contour and corner point extraction

下载: 全尺寸图片幻灯片

图 5 车道线虚线角点提取试验结果

Figure 5. Extraction test results of the dotted line corners of the roadway

下载: 全尺寸图片幻灯片

图 6 车道线虚线角点提取流程

Figure 6. Flowchart for extracting the dotted line corner points of the roadway

下载: 全尺寸图片幻灯片

图 7 车道线虚线角点成像示意图

Figure 7. Schematic diagram of imaging of dotted corners of roadways

下载: 全尺寸图片幻灯片

图 8 试验路段图像示例

Figure 8. Picture of the test roads

下载: 全尺寸图片幻灯片

图 9 与文献[7]~[8]的结果对比

Figure 9. Comparison results with methods in references [7]~[8]

下载: 全尺寸图片幻灯片

图 10 参数录入信息

Figure 10. Import of parameters

下载: 全尺寸图片幻灯片

图 11 视距计算结果输出

Figure 11. Output of sight distance calculation result

下载: 全尺寸图片幻灯片

表 1 二、三、四级公路的停车、会车和超车视距

Table 1. Sight distance for parking, meeting and overtaking on second, third, and fourth level highways

设计速度/(km/h)	停车视距/m	会车视距/m	超车视距/m
80	110	220	550
60	75	150	350
40	40	80	200

下载: 导出CSV

表 2 计算误差

Table 2. Calculation error

序号	测量值/m	计算值/m	绝对误差/m	相对误差/%
1	24.2	25.51	1.31	5.41
2	17.4	18.14	0.74	4.52
3	20.2	21.34	1.14	5.64
4	25.3	27.01	1.71	6.76
5	23.3	24.92	1.62	6.95
6	29.8	31.76	1.96	6.58
7	18.5	19.19	0.69	3.73
8	21.3	22.61	1.31	6.15

下载: 导出CSV

参考文献(20)

[1]	张卫华, 张鑫, 曹世全, 等. 适应能见度变化的道路线形诱导标志设置方法[J]. 中国安全科学学报, 2019, 29(7): 76-83. https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201907017.htm ZHANG W H, ZHANG X, CAO S Q, et al. Road alignment induction sign setting method adapting to visibility change[J]. China Safety Science Journal, 2019, 29(7): 76-83. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201907017.htm
[2]	文森, 梁波, 肖尧, 等. 基于反应时间的公路隧道接近段停车视距研究[J]. 交通信息与安全, 2021, 39(2): 43-52. doi: 10.3963/j.jssn.1674-4861.2021.02.006 WEN S, LIANG B, XIAO Y, et al. A stopping distance in access zone of highway tunnel based on reaction time[J]. Journal of Transport Information and Safety, 2021, 39(2): 43-52 (in Chinese) doi: 10.3963/j.jssn.1674-4861.2021.02.006
[3]	WANG Y, TEOH K, SHEN D G. Lane detection and tracking using B-Snake[J]. Image and Vision Computing. 2004, 22 (4): 269-280. doi: 10.1016/j.imavis.2003.10.003
[4]	郭磊, 李克强, 王建强, 等. 应用方向可调滤波器的车道线识别方法[J]. 机械工程学报, 2008(8): 214-218+226. doi: 10.3321/j.issn:0577-6686.2008.08.038 GUO L, LI K Q, WANG J Q, et al. Lane detection method by steerable filters[J]. Chinese Journal of Mechanical Engineering, 2008(8): 214-218+226. (in Chinese) doi: 10.3321/j.issn:0577-6686.2008.08.038
[5]	ANDRDE C, BUENO F, FRANCO R, et al. A novel strategy for road lane detection and tracking based on a vehicle's forward monocular camera[J]. IEEE Transactions on Intelligent Transportation Systems, 2018, 20(4): 1497-1507.
[6]	XING Y, LYU C, WANG H J, et al. Dynamic integration and online evaluation of vision-based lane detection algorithms[J]. IET Intelligent Transport Systems, 2018, 13(1): 55-62.
[7]	ZHANG Z Y, QIAN W, PAN L, et al. Adaptive zoom distance measuring system of camera based on the ranging of binocular vision[J]. Modern Applied Science, 2012, 6(5): 43-47.
[8]	高波. 基于单目视觉的公路视距检测技术研究[D]. 西安: 长安大学, 2017. GAO B. Research on road slight distance detection technology based on monocular vision[D]. Xi'an: Chang'an University, 2017. (in Chinese)
[9]	白琛琛. 高等级公路弯道视距测算及安全评价方法研究[D]. 西安: 长安大学, 2020. BAI C C. Research on sight distance calculation and safety evaluation method for high-grade highway curves[D]. Xi'an: Chang'an University, 2020.(in Chinese)
[10]	CHEN Y, HE M Y. Sharp curve lane boundaries projective model and detection[C]. 10^th International Conference on Industrial Informatics, Beijing, China: IEEE, 2012.
[11]	AGRAWAL S, DEO K, HALDAR S, et al. Off-road lane detection using superpixel clustering and ransac curve fitting[C]. 15^th International Conference on Control, Automation, Robotics and Vision(ICARCV), Singapore: IEEE, 2018
[12]	SCHOMERUS V, ROSEBROCK D, WAHL M. Camera-based lane border detection in arbitrarily structured environments[C]. 2014 IEEE Intelligent Vehicles Symposium, Michigan, USA: IEEE, 2014.
[13]	PARK Y, HWANG Y. Robust range estimation with a monocular camera for vision-based forward collision warning system[J]. The Scientific World Journal, 2014(12): 1-9.
[14]	陈雨人, 付云天, 汪凡. 基于支持向量回归的视距计算模型建立和应用[J]. 中国公路学报, 2018, 31(4): 105-113. doi: 10.3969/j.issn.1001-7372.2018.04.013 CHEN Y, FU Y T, WANG F. Establishment and application of slight distance computing model based on support vector regression[J]. China Journal of Highway and Transport, 2018, 31(4): 105-113. doi: 10.3969/j.issn.1001-7372.2018.04.013
[15]	LEBMANN S, MEUTER M, MULLER D, et al. Probabilistic distance estimation for vehicle tracking application in monocular vision[C]. 4^th IEEE Intelligent Vehicles Symposium, Gothenburg, Sweden: IEEE, 2016.
[16]	JWP A, YWH A, YZY A, et al. Development of an embedded road boundary detection system based on deep learning[J]. Image and Vision Computing, 2020, 100(5), 82-95.
[17]	曹思佳, 代扬, 余洪山, 等. 基于机器视觉的机械表走时精度测量[J]. 湖南大学学报(自然科学版), 2020, 47(12): 86-94. https://www.cnki.com.cn/Article/CJFDTOTAL-HNDX202012011.htm CAO S J, DAI Y, YU H S, et al. Accuracy measurement of mechanical watch travel time based on machine vision[J]. Journal of Hunan University(Natural Science), 2020, 47 (12): 86-94. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HNDX202012011.htm
[18]	钱文光, 林小竹. 基于轮廓尖锐度的图像角点检测算法[J]. 计算机工程, 2008(6): 202-204. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJC200806075.htm QIAN W G, LIN X Z. Detection algorithm of image corner based contour sharp degree[J]. Computer Engineering, 2008 (6): 202-204.(in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JSJC200806075.htm
[19]	刘燕. 基于抽样和最大最小距离法的并行K-means聚类算法[J]. 智能计算机与应用, 2018, 8(6): 37-39+43. doi: 10.3969/j.issn.2095-2163.2018.06.008 LIU Y. Parallel K-means clustering algorithm based on sampling and maximum & minimum distance method[J]. Intelligent Computer and Applications, 2018, 8(6): 37-39+43. (in Chinese) doi: 10.3969/j.issn.2095-2163.2018.06.008
[20]	郑榜贵, 田炳香, 段建民. 基于交比不变量的摄像机标定方法[J]. 北京工业大学学报, 2008(5): 476-480. https://www.cnki.com.cn/Article/CJFDTOTAL-BJGD200805008.htm ZHEN B G, TIAN B X, DUAN J M. Camera calibration approach based on cross-ratio invariability[J]. Journal of Beijing University of Technology, 2008(5): 476-480. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-BJGD200805008.htm