基于单幅立式标靶图像的单目深度信息提取

黄小云; 高峰; 徐国艳; 丁能根; 邢龙龙

doi:10.13700/j.bh.1001-5965.2014.0272

基于单幅立式标靶图像的单目深度信息提取

doi: 10.13700/j.bh.1001-5965.2014.0272

北京航空航天大学交通科学与工程学院, 北京 100191

基金项目: 国家自然科学基金资助项目(51105021); 国家自然科学基金青年基金资助项目(51405008); 北京市自然科学基金资助项目(3133040)

详细信息

作者简介:
黄小云(1985—),女,湖南邵阳人,博士生,hxly0820@126.com

通讯作者:
高峰(1955—),男,河南禹州人,教授,gaof@buaa.edu.cn,主要研究方向为智能车辆.

中图分类号: U491
计量
- 文章访问数: 1260
- HTML全文浏览量: 131
- PDF下载量: 809
- 被引次数: 0
出版历程
- 收稿日期: 2014-05-14
- 修回日期: 2014-07-25
- 网络出版日期: 2015-04-20

Depth information extraction of on-board monocular vision based on a single vertical target image

School of Transportation Science and Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100191, China

摘要

摘要: 针对在智能车中车载单目视觉系统已经检测到路面障碍物的情况下,计算障碍物相对于本车的距离问题,提出了一种仅利用单幅标靶图像且无需相机内部参数的图像深度信息提取方法．该方法利用一种放置于相机前方的立式标靶,建立图像纵坐标像素与实际成像角度之间的映射关系,结合投影几何模型实现实时深度信息的提取．依据立式标靶图像的特点,设计了包括标靶图像感兴趣区设置、模板匹配、候选点聚类、筛选及精确定位等处理的亚像素级角点检测及定位算法．实验结果显示,该方法具有较高的测量精度及实时性．相对于在路面摆放参照物的方法,该方法无需大标定场地且规避了数据拟合引起的误差．同时该方法标定只需一幅图像,过程简单,便于实际应用．
- 机器视觉 /
- 单目测距 /
- 立式标靶 /
- 智能车 /
- 角点检测
Abstract: Calculating obstacle distance for on-board monocular vision system on intelligent vehicle was investigated when the obstacle has been detected. An extraction method of depth information only using a single target image without any internal camera parameters was developed. The mapping relation between image row pixel values and the actual imaging angles was established with the image of vertical target, which was placed in the front of a camera. The obstacle depth information was extracted in real time by combining the projection geometry model. Given the characteristic of vertical target image, an algorithm of sub-pixel corner detection and location was designed, which includes region of interest setting, template matching, candidate points clustering and screening and precise location, etc. Experimental results show that the method has high precision and real-time performance. Compared with the method of putting reference on the road, it does not need large calibration site and could avoid the data fitting error. And the method also has a simple calibration procedure with a single image, which is suitable for practical application.
- machine vision /
- monocular distance detection /
- vertical target /
- intelligent vehicle /
- corner detection

HTML全文

参考文献(16)

[1]	Xu F F, Wang B,Zhou Z Q,et al.Real-time lane detection for intelligent vehicles based on monocular vision[C]//Proceedings of the 31st Chinese Control Conference.Washington D.C.:IEEE,2012:7332-7337.
[2]	Sivaraman S, Trivedi M M.Combining monocular and stereo-vision for real-time vehicle ranging and tracking on multilane highways[C]//Proceedings of International IEEE Conference on Intelligent Transportation Systems.Washington D.C.:IEEE,2011:1249-1254.
[3]	王浩,许志闻, 谢坤,等.基于OpenCV的双目测距系统[J].吉林大学学报,2014,32(2):188-194. Wang H,Xu Z W,Xie K,et al.Binocular measuring system based on OpenCV[J].Journal of Jilin University,2014,32(2):188-194(in Chinese).
[4]	Sun W, Chen L,Hu B,et al.Binocular vision-based position determination algorithm and system[C]//Proceedings of the 2012 International Conference on Computer Distributed Control and Intelligent Environmental Monitoring.Piscataway:IEEE Computer Society,2012:170-173.
[5]	于乃功,黄灿, 林佳.基于单目视觉的机器人目标定位测距方法研究[J].计算机测量与控制,2012,20(10):2654-2660. Yu N G,Huang C,Lin J.Target location and ranging based on monocular vision robot[J].Computer Measurement & Control,2012,20(10):2654-2660(in Chinese).
[6]	吴刚,唐振民. 单目式自主机器人视觉导航中的测距研究[J].机器人,2010,32(6):828-832. Wu G,Tang Z M.Distance measurement in visual navigation of monocular autonomous robots[J].Robot,2010,32(6):828-832(in Chinese).
[7]	鲁威威,肖志涛, 雷美琳.基于单目视觉的前方车辆检测与测距方法研究[J].视频应用与工程,2011,35(1):125-128. Lu W W,Xiao Z T,Lei M L.Study on method of preceding vehicles detection and distance measurement based on monocular vision[J].Video Application & Project,2011,35(1):125-128(in Chinese).
[8]	战强,吴佳. 未知环境下移动机器人单目视觉导航算法[J].北京航空航天大学学报,2008,34(6):613-617. Zhan Q,Wu J.Monocular vision-based navigation algorithm for mobile robots in unknown environments[J].Journal of Beijing University of Aeronautics and Astronautics,2008,34(6):613-617(in Chinese).
[9]	王青. 基于机器视觉的车辆距离测量技术[D].南京:南京理工大学,2008. Wang Q.Vehicle distance measurement technology based on machine vision[D].Nanjing:Nanjing University of Science and Technology,2008(in Chinese).
[10]	Wu C F, Lin C J,Lee C Y,et al.Applying a functional neurofuzzy network to real-time lane detection and front-vehicle distance measurement[J].IEEE Transactions on Systems,Man,and Cybernetics-Part C:Applications and Reviews,2012,42(4): 577-589.
[11]	Wu K,Di K C, Sun X,et al.Enhanced monocular visual odometry integrated with laser distance meter for astronaut navigation[J].Sensors,2014,14(3):4981-5003.
[12]	王荻,王洁,付小宁. 基于单目图像和方向的测距系统及性能分析[J].光子学报,2011,40(7):1112-1116. Wang D,Wang J,Fu X N.A passive ranging system based on image sequence from single lens and the respective imaging direction along with its performance[J].Acta Photonica Sinica,2011,40(7):1112-1116(in Chinese).
[13]	赵信宇,朱晓蕊, 余锦全.单幅图像深度信息的提取[J].制造业自动化,2010,32(3):15-17. Zhao X Y,Zhu X R,Yu J Q.Learning depth information from monocular image[J].Manufacturing Automation,2010,32(3): 15-17(in Chinese).
[14]	韩延祥,张志胜, 戴敏.用于目标测距的单目视觉测量方法[J].光学精密工程,2011,19(5):1110-1117. Han Y X,Zhang Z S,Dai M.Monocular vision system for distance measurement based on feature points[J].Optics and Precision Engineering,2011,19(5):1110-1117(in Chinese).
[15]	Otsu N. A threshold selection method from gray-level histograms[J].IEEE Transactions on Systems,Man,and Cybernetics,1979,9(1):62-66.
[16]	刘瑞祯,于仕琪. OpenCV教材[M].北京:北京航空航天大学出版社,2007:249-250. Liu R Z,Yu S Q.OpenCV teaching material[M].Beijing:Beihang University Press,2007:249-250(in Chinese).