双目测距技术在调车作业中的研究与应用
|
摘要
为实现调车作业视频监控与测距的目的,设计一种用于调车顶送作业的双目测距系统,其通过安装于机车车厢的双目摄像机,采集机车已挂车厢与待挂车厢之间实时视频并测量其距离。在OpenCV计算机视觉库与Windows平台下,使用张正友标定法对双目摄像机进行标定,并获得畸变校正后的图像与摄像机参数。通过SGBM特征匹配算法实现车厢目标特征提取与立体匹配。最后利用三维重建方法获得深度信息。系统分别在实验室环境与现场环境下进行实验,现场实验结果表明,测距结果比实际距离稍偏大,最大误差小于1 m,满足实际现场环境要求,可有效地辅助人工瞭望进行调车作业。
In order to achieve the purpose of video surveillance and distance measurement in shunting operation, a distance measurement system was designed. With a camera mounted on the connected carriage, it collected the real-time video and measured distance among carriage connected and unconnected. Under the OpenCV computer vision and windows platform, the intrinsic and extrinsic parameters of cameras and rectified images were determined using Zhang's calibration method. The target feature extraction and stereo matching were achieved though the SGBM feature matching algorithm. Finally, the depth information was obtained using 3D-reconstruction method. The distance measurement system has been tested under the laboratory environment and field environment. The experimental results show that the distance of the system is slightly larger than the actual one, and the maximum error is less than 1 meters. It can effectively assist manual work for shunting operation.
In order to achieve the purpose of video surveillance and distance measurement in shunting operation, a distance measurement system was designed. With a camera mounted on the connected carriage, it collected the real-time video and measured distance among carriage connected and unconnected. Under the OpenCV computer vision and windows platform, the intrinsic and extrinsic parameters of cameras and rectified images were determined using Zhang's calibration method. The target feature extraction and stereo matching were achieved though the SGBM feature matching algorithm. Finally, the depth information was obtained using 3D-reconstruction method. The distance measurement system has been tested under the laboratory environment and field environment. The experimental results show that the distance of the system is slightly larger than the actual one, and the maximum error is less than 1 meters. It can effectively assist manual work for shunting operation.
引文
[1]刘鹏鑫.铁路调车人因事故可靠性研究[J].铁道货运,2016, 34(8):45-49.LIU Pengxin. Study of the reliability of human factor in railway shunting accident[J]. Railway Freight Transport,2016, 34(8):45-49.
[2]张亮.铁路车务系统中间站调车作业安全防控对策分析[J].铁道运输与经济, 2018, 40(5):71-74.ZHANG Liang. An analysis on the safety management of shunting operation at intermediate stations[J]. Railway Transport and Economy, 2018, 40(5):71-74.
[3]曹海霞.无线调车机车信号和监控系统(STP)的应用分析[J].铁道通信信号, 2017, 53(9):33-36.CAO Haixia. Application analysis of STP[J]. Railway Signalling&Communication, 2017, 53(9):33-36.
[4]龚骁猷.铁路GSM-R网络编号方案关键问题研究[J].铁道标准设计, 2018, 62(2):160-166.GONG Xiaoyou. Research on key problems of railway GSM-R network numbering program[J]. Railway Standard Design, 2018, 62(2):160-166.
[5]黄铁军,郑锦,李波,等.多媒体技术研究:2013——面向智能视频监控的视觉感知与处理[J].中国图象图形学报, 2014, 19(11):1539-1562.HUANG Tiejun, ZHENG Jin, LI Bo, et al. Visual perception and processing for intelligent video surveillance:a review[J]. Journal of Image and Graphics,2014, 19(11):1539-1562.
[6]黄松梅,毕远伟,刘殿通,等.双目立体视觉非接触式测量研究[J].烟台大学学报(自然科学与工程版), 2017,30(4):323-327.HUANG Songmei, BI Yuanwei, LIU Diantong, et al.Non-contact measurement with binocular stereo vision[J].Journal of Yantai University(Natural Science and Engineering Edition), 2017, 30(4):323-327.
[7]安果维,王耀南,周显恩,等.基于显著性检测的双目测距系统[J].智能系统学报, 2018(6):1-9.AN Guowei, WANG Yaonan, ZHOU Xian’en, et al.Binocular distance measurement system based on saliency detection[J]. CAAI Transactions on Intelligent Systems, 2018(6):1-9.
[8] ZHANG Zhengyou. A flexible new technique for camera cablibration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(11):1330-1334.
[9]许威,丁学文,宋兰草,等.基于MATLAB和OpenCV的双目视觉测距系统的实现[J].天津职业技术师范大学学报, 2017, 27(4):47-52.XU Wei, DING Xuewen, SONG Lancao, et al.Realization of binocular stereo vision distance measurement system based on MATLAB and OpenCV[J].Journal of Tianjin University of Technology and Education, 2017, 27(4):47-52.
[10]汪珍珍,赵连玉,刘振忠.基于MATLAB与OpenCV相结合的双目立体视觉测距系统[J].天津理工大学学报, 2013, 29(1):45-48.WANG Zhenzhen, ZHAO Lianyu, LIU Zhenzhong.Binocular stereo vision distance measurement system based on a combination of Matlab and OpenCV[J].Journal of Tianjin University of Technology, 2013, 29(1):45-48.
[11]张欢,安利,张强,等. SGBM算法与BM算法分析研究[J].测绘与空间地理信息, 2016, 39(10):214-216.ZHANG Huan, AN Li, ZHANG Qiang, et al. SGBM algorithm and B-M algorithm analysis and research[J].Geomatics&Spatial Information Technology, 2016,39(10):214-216.
[12] Rostam H, Haidi I, Eduard L. Literature survey on stereo vision disparity map algorithms[J]. Journal of Sensors,2015, 10(2):34-51.
[13] Hirschmuller H. Accurate and efficient stereo processing by semi global matching and mutual information[C]//Proc of IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE Press, 2005:807-814.
[14] Bleyer M, Gelautz M. Graph-cut-based stereo matching using image segmentation with symmetrical treatment of occlusions[J]. Signal Processing:Image Communication,2007, 22(2):127-143.
[15]杨韬,符文星,王民钢,等.基于光场成像的双目深度图像获取[J].西北工业大学学报, 2015, 33(5):727-731.YANG Tao, FU Wenxing, WANG Minggang, et al.Binocular stereo depth map acquisition based on light field images[J]. Journal of Northwestern Polytechnical University, 2015, 33(5):727-731.
[16]任继昌,杨晓东.基于互信息的高精度双目视觉测距方法研究[J].控制工程, 2015, 22(1):199-204.REN Jichang, YANG Xiaodong. Improving rangemeasuring accuracy of stereo vision based on mutual information registration[J]. Control Engineering of China,2015, 22(1):199-204.
[17]王予见.图像处理技术在调车作业安全防护中的应用研究[D].兰州:兰州交通大学, 2015.WANG Yujian. Application research of image processing technologies on safety protection for shunting operation[D]. Lanzhou:Lanzhou Jiaotong University, 2015.
[18]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 Zhuoyuan. The design of preceding vehicles detection and distance measurement system based on machine vision[D]. Harbin:Harbin Institute of Technology, 2015.
[19]张颖江,潘尧,吴聪.基于车载系统双目CCD相机测距[J].信息安全与技术, 2016, 7(1):57-62.ZHANG Yinjiang, PAN Yao, WU Cong. Distance measurement of binocular CCD camera on vehicle-mounted system[J]. Information Security and Technology, 2016, 7(1):57-62.
[20]王志远,王茂森.基于双目视觉的移动机器人测距与定位系统[J].兵器装备工程学报, 2017, 38(11):173-177.WANG Zhiyuan, WANG Maosen. Ranging and positioning system of mobile robot based on binocular vision[J]. Journal of Ordnance Equipment Engineering,2017, 38(11):173-177.
[2]张亮.铁路车务系统中间站调车作业安全防控对策分析[J].铁道运输与经济, 2018, 40(5):71-74.ZHANG Liang. An analysis on the safety management of shunting operation at intermediate stations[J]. Railway Transport and Economy, 2018, 40(5):71-74.
[3]曹海霞.无线调车机车信号和监控系统(STP)的应用分析[J].铁道通信信号, 2017, 53(9):33-36.CAO Haixia. Application analysis of STP[J]. Railway Signalling&Communication, 2017, 53(9):33-36.
[4]龚骁猷.铁路GSM-R网络编号方案关键问题研究[J].铁道标准设计, 2018, 62(2):160-166.GONG Xiaoyou. Research on key problems of railway GSM-R network numbering program[J]. Railway Standard Design, 2018, 62(2):160-166.
[5]黄铁军,郑锦,李波,等.多媒体技术研究:2013——面向智能视频监控的视觉感知与处理[J].中国图象图形学报, 2014, 19(11):1539-1562.HUANG Tiejun, ZHENG Jin, LI Bo, et al. Visual perception and processing for intelligent video surveillance:a review[J]. Journal of Image and Graphics,2014, 19(11):1539-1562.
[6]黄松梅,毕远伟,刘殿通,等.双目立体视觉非接触式测量研究[J].烟台大学学报(自然科学与工程版), 2017,30(4):323-327.HUANG Songmei, BI Yuanwei, LIU Diantong, et al.Non-contact measurement with binocular stereo vision[J].Journal of Yantai University(Natural Science and Engineering Edition), 2017, 30(4):323-327.
[7]安果维,王耀南,周显恩,等.基于显著性检测的双目测距系统[J].智能系统学报, 2018(6):1-9.AN Guowei, WANG Yaonan, ZHOU Xian’en, et al.Binocular distance measurement system based on saliency detection[J]. CAAI Transactions on Intelligent Systems, 2018(6):1-9.
[8] ZHANG Zhengyou. A flexible new technique for camera cablibration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(11):1330-1334.
[9]许威,丁学文,宋兰草,等.基于MATLAB和OpenCV的双目视觉测距系统的实现[J].天津职业技术师范大学学报, 2017, 27(4):47-52.XU Wei, DING Xuewen, SONG Lancao, et al.Realization of binocular stereo vision distance measurement system based on MATLAB and OpenCV[J].Journal of Tianjin University of Technology and Education, 2017, 27(4):47-52.
[10]汪珍珍,赵连玉,刘振忠.基于MATLAB与OpenCV相结合的双目立体视觉测距系统[J].天津理工大学学报, 2013, 29(1):45-48.WANG Zhenzhen, ZHAO Lianyu, LIU Zhenzhong.Binocular stereo vision distance measurement system based on a combination of Matlab and OpenCV[J].Journal of Tianjin University of Technology, 2013, 29(1):45-48.
[11]张欢,安利,张强,等. SGBM算法与BM算法分析研究[J].测绘与空间地理信息, 2016, 39(10):214-216.ZHANG Huan, AN Li, ZHANG Qiang, et al. SGBM algorithm and B-M algorithm analysis and research[J].Geomatics&Spatial Information Technology, 2016,39(10):214-216.
[12] Rostam H, Haidi I, Eduard L. Literature survey on stereo vision disparity map algorithms[J]. Journal of Sensors,2015, 10(2):34-51.
[13] Hirschmuller H. Accurate and efficient stereo processing by semi global matching and mutual information[C]//Proc of IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE Press, 2005:807-814.
[14] Bleyer M, Gelautz M. Graph-cut-based stereo matching using image segmentation with symmetrical treatment of occlusions[J]. Signal Processing:Image Communication,2007, 22(2):127-143.
[15]杨韬,符文星,王民钢,等.基于光场成像的双目深度图像获取[J].西北工业大学学报, 2015, 33(5):727-731.YANG Tao, FU Wenxing, WANG Minggang, et al.Binocular stereo depth map acquisition based on light field images[J]. Journal of Northwestern Polytechnical University, 2015, 33(5):727-731.
[16]任继昌,杨晓东.基于互信息的高精度双目视觉测距方法研究[J].控制工程, 2015, 22(1):199-204.REN Jichang, YANG Xiaodong. Improving rangemeasuring accuracy of stereo vision based on mutual information registration[J]. Control Engineering of China,2015, 22(1):199-204.
[17]王予见.图像处理技术在调车作业安全防护中的应用研究[D].兰州:兰州交通大学, 2015.WANG Yujian. Application research of image processing technologies on safety protection for shunting operation[D]. Lanzhou:Lanzhou Jiaotong University, 2015.
[18]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 Zhuoyuan. The design of preceding vehicles detection and distance measurement system based on machine vision[D]. Harbin:Harbin Institute of Technology, 2015.
[19]张颖江,潘尧,吴聪.基于车载系统双目CCD相机测距[J].信息安全与技术, 2016, 7(1):57-62.ZHANG Yinjiang, PAN Yao, WU Cong. Distance measurement of binocular CCD camera on vehicle-mounted system[J]. Information Security and Technology, 2016, 7(1):57-62.
[20]王志远,王茂森.基于双目视觉的移动机器人测距与定位系统[J].兵器装备工程学报, 2017, 38(11):173-177.WANG Zhiyuan, WANG Maosen. Ranging and positioning system of mobile robot based on binocular vision[J]. Journal of Ordnance Equipment Engineering,2017, 38(11):173-177.