面向智能交通的单目视觉测距方法研究
|
摘要
与前方车辆距离是影响行车安全的重要因素,因此本文提出一种面向未来智能交通的前方车辆单目视觉测距方法.首先,提出融合物联网、智能识别、云计算技术的车联网模型,车辆可实时向车联网回传位置信息及前车图像,请求附近交通标志及前方车辆几何尺度信息,车辆端可计算图像坐标系下车道标志线、交通标志、车辆尺度信息.然后,建立单目相机数学模型,介绍以交通标志、车道分界线为合作标志的单目视觉测距方法.最后,综合应用单目视觉测距方法,设计了前方车辆自适应视觉测距方案.通过仿真实验,证明了单目视觉测距方法的正确性与有效性,可丰富驾驶辅助系统的前方车辆测距手段.
The preceding vehicle distance is a significant factor, affecting driving safety. A preceding vehicle distance measurement based on monocular vision for future intelligent traffic system is proposed. First, the model of internet of vehicles is proposed with the fusion of internet of things, intelligent recognition, and cloud computing technology. The vehicle can send back the location information and the preceding vehicle image to the internet of vehicles in real time, and request the nearby traffic signs and the geometry of the preceding vehicle.Then, establish a mathematical model of the monocular camera, and introduce a distance measurement based on monocular vision with a cooperation sign of traffic signs and lane lines. Finally, a preceding vehicle adaptive vision distance measurement is designed by comprehensive application of the distance measurement based on monocular vision. The simulation demonstrates that the distance measurement based on monocular vision is valid and effective, enriching driving assistance system.
The preceding vehicle distance is a significant factor, affecting driving safety. A preceding vehicle distance measurement based on monocular vision for future intelligent traffic system is proposed. First, the model of internet of vehicles is proposed with the fusion of internet of things, intelligent recognition, and cloud computing technology. The vehicle can send back the location information and the preceding vehicle image to the internet of vehicles in real time, and request the nearby traffic signs and the geometry of the preceding vehicle.Then, establish a mathematical model of the monocular camera, and introduce a distance measurement based on monocular vision with a cooperation sign of traffic signs and lane lines. Finally, a preceding vehicle adaptive vision distance measurement is designed by comprehensive application of the distance measurement based on monocular vision. The simulation demonstrates that the distance measurement based on monocular vision is valid and effective, enriching driving assistance system.
引文
[1]李祎承,胡钊政,胡月志,等.基于GPS与图像融合的智能车辆高精度定位算法[J].交通运输系统工程与信息,2017,17(3):112-119.[LI Y C,HU Z Z,HU Y Z,et al.Accurate localization based on GPS and image fusion for intelligent vehicles[J].Journal of Transportation Systems Engineering and Information Technology,2017,17(3):112-119.]
[2]覃华.大数据驱动的车联网信息物理融合系统的分析与设计方法[D].广州:广东工业大学,2016.[QIN HThe analysis and design method of big data-driven vehicular network cyber-physical system[D]Guangzhou:Guangdong Institute of Technology,2016.]
[3]钟晓明,余贵珍,马亚龙,等.基于快速区域卷积神经网络的交通标志识别算法研究[C].2016中国汽车工程学会年会论文集,2016.[ZHONG X M,YU G Z,MAY L,et al.Traffic sign recognition based on fast regionbased convolutional neural networks[C].2016 CG-IVT2016.]
[4]曹诗雨,刘跃虎,李辛昭.基于Fast R-CNN的车辆目标检测[J].中国图象图形学报,2017,22(5):671-677[CAO S Y,LIU Y H,LI X Z.Vehicle detection method based on fast R-CNN[J].Journal of Image and Graphics2017,22(5):671-677.]
[5]王乐芬.基于单目视觉的多特征前方车辆检测及测距方法研究[D].秦皇岛:燕山大学,2016.[WANG L FStudy on preceding vehicles detection with multifeatures and distance measurement based on monocular vision[D].Qinhuangdao:Yanshan University,2016.]
[6]佟卓远.基于机器视觉的前方车辆检测与测距系统设计[D].哈尔滨:哈尔滨工业大学,2015.[TONG Z Y.The design of preceding vehicles detection and distance measurement system based on machine vision[D].Harbin:Harbin Institute of Technology,2015.]
[7]徐德,谭民,李原.机器人视觉测量与控制(第2版)[M].北京:国防工业出版社,2011.[XU D,TAN M,LI Y,Visual measurement and control for robot(2nd edition)[M].Beijing:National Defense Industry Press,2011.]
[8]XU L,CAO Z,LIU X.A monocular vision system for pose measurement in indoor environment[C].IEEEInternational Conference on Robotics and Biomimetics,IEEE,2017:1977-1982.
[2]覃华.大数据驱动的车联网信息物理融合系统的分析与设计方法[D].广州:广东工业大学,2016.[QIN HThe analysis and design method of big data-driven vehicular network cyber-physical system[D]Guangzhou:Guangdong Institute of Technology,2016.]
[3]钟晓明,余贵珍,马亚龙,等.基于快速区域卷积神经网络的交通标志识别算法研究[C].2016中国汽车工程学会年会论文集,2016.[ZHONG X M,YU G Z,MAY L,et al.Traffic sign recognition based on fast regionbased convolutional neural networks[C].2016 CG-IVT2016.]
[4]曹诗雨,刘跃虎,李辛昭.基于Fast R-CNN的车辆目标检测[J].中国图象图形学报,2017,22(5):671-677[CAO S Y,LIU Y H,LI X Z.Vehicle detection method based on fast R-CNN[J].Journal of Image and Graphics2017,22(5):671-677.]
[5]王乐芬.基于单目视觉的多特征前方车辆检测及测距方法研究[D].秦皇岛:燕山大学,2016.[WANG L FStudy on preceding vehicles detection with multifeatures and distance measurement based on monocular vision[D].Qinhuangdao:Yanshan University,2016.]
[6]佟卓远.基于机器视觉的前方车辆检测与测距系统设计[D].哈尔滨:哈尔滨工业大学,2015.[TONG Z Y.The design of preceding vehicles detection and distance measurement system based on machine vision[D].Harbin:Harbin Institute of Technology,2015.]
[7]徐德,谭民,李原.机器人视觉测量与控制(第2版)[M].北京:国防工业出版社,2011.[XU D,TAN M,LI Y,Visual measurement and control for robot(2nd edition)[M].Beijing:National Defense Industry Press,2011.]
[8]XU L,CAO Z,LIU X.A monocular vision system for pose measurement in indoor environment[C].IEEEInternational Conference on Robotics and Biomimetics,IEEE,2017:1977-1982.