面向低视角场面监视的移动目标速度测量

张天慈; 丁萌; 钱小燕; 左洪福

doi:10.13700/j.bh.1001-5965.2019.0234

面向低视角场面监视的移动目标速度测量

doi: 10.13700/j.bh.1001-5965.2019.0234

1.
南京林业大学汽车与交通工程学院, 南京 210037
2.
南京航空航天大学民航学院, 南京 211106

基金项目:

国家自然科学基金委员会-中国民用航空局民航联合研究基金 U1633105

国家自然科学基金 61803199

航空科学基金 20170752008

江苏省高等学校自然科学研究面上项目 19KJB580013

详细信息

作者简介:
张天慈男，博士，讲师。主要研究方向：智能交通与新航行系统。E-mail:tczhang@njfu.edu.cn

丁萌男，博士，副教授，硕士生导师。主要研究方向：机场场面监控，无人机导航、制导与控制，民用飞机航电系统适航技术

钱小燕女，博士，副教授，硕士生导师。主要研究方向：智能交通、机场场面监视技术、图像处理与分析

通讯作者:
张天慈. E-mail:tczhang@njfu.edu.cn

中图分类号: V351.11
计量
- 文章访问数: 836
- HTML全文浏览量: 215
- PDF下载量: 148
- 被引次数: 0
出版历程
- 收稿日期: 2019-05-18
- 录用日期: 2019-07-05
- 网络出版日期: 2020-02-20

Moving object speed measurement for low-camera-angle surface surveillance

1.
College of Automobile and Traffic Engineering, Nanjing Forestry University, Nanjing 210037, China
2.
College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China

Funds:

Joint Research Funds of National Natural Science Foundation of China and Civil Aviation Administration of China U1633105

National Natural Science Foundation of China 61803199

Aeronautical Science Foundation of China 20170752008

Natural Science Foundation of Jiangsu Higher Education Institutions of China 19KJB580013

More Information

Corresponding author: ZHANG Tianci, E-mail:tczhang@njfu.edu.cn

摘要

摘要:
为构建有效的机场场面视觉监视系统，提出了一种基于特征点持续跟踪与分析的移动目标速度测量方法。首先，利用场面几何特征对摄像机进行标定；然后，基于光流场对图像运动区域的特征点进行持续跟踪，在此基础上通过特征点轨迹聚类区分不同移动目标；最后，根据特征点高度与运动距离完成速度测量。所提方法能够利用机场场面摄像机获取的低视角单目视频图像，对移动目标的运动速度进行准确测量。基于广州白云国际机场的场面运行视频进行了仿真分析，验证了所提方法在低视角速度测量方面的可行性与优势。
- 机场场面监视 /
- 低视角 /
- 轨迹聚类 /
- 速度测量 /
- 标定
Abstract:
To build an effective airport surface visual surveillance system, a moving object speed measurement method based on long-term feature point tracking and analysis is proposed. First, the surveillance camera is calibrated using geographic features on the airport surface. Then, the feature points in motion regions of the images are continuously tracked via optical flow fields. On this basis, different moving objects are identified by clustering the feature point trajectories. Finally, the speeds of the moving objects are measured according to the heights and moving distances of the feature points. The proposed method can accurately measure the object moving speeds using low-camera-angle monocular video images obtained by cameras installed on the airport surface. Simulation studies are conducted based on the surface operation videos of Guangzhou Baiyun International Airport, which verify the feasibility and advantages of the proposed method for low-camera-angle speed measurement.
- airport surface surveillance /
- low camera angle /
- trajectory clustering /
- speed measurement /
- calibration

HTML全文

图 1 系统方案框图

Figure 1. Block diagram of system scheme

下载: 全尺寸图片幻灯片

图 2 空间坐标系与图像坐标系位置关系示意图

Figure 2. Illustration of position relation between spatial coordinate system and image coordinate system

下载: 全尺寸图片幻灯片

图 3 用于摄像机标定的场面特征

Figure 3. Surface features used for camera calibration

下载: 全尺寸图片幻灯片

图 4 摄像机标定结果图示

Figure 4. Illustration of camera calibration results

下载: 全尺寸图片幻灯片

图 5 特征点轨迹聚类

Figure 5. Feature point trajectory clustering

下载: 全尺寸图片幻灯片

图 6 机场平面轨迹投影

Figure 6. Trajectory projection on airport surface

下载: 全尺寸图片幻灯片

图 7 移动目标速度测量结果

Figure 7. Moving object speed measurement results

下载: 全尺寸图片幻灯片

图 8 速度测量误差对比

Figure 8. Comparison of speed measurement errors

下载: 全尺寸图片幻灯片

表 1 场面运行视频数据集

Table 1. Dataset of surface operation videos

视频名称	分辨率	帧数	航空器个数	车辆个数
cam2513-视频1	1024×576	401	1	0
cam2513-视频2	1024×576	101	1	0
cam2513-视频3	1024×576	574	1	4
cam9915-视频1	1024×576	801	1	1
cam9915-视频2	1024×576	451	0	2
cam9915-视频3	1024×576	551	2	1

下载: 导出CSV

表 2 摄像机标定结果

Table 2. Camera calibration results

摄像机	消失点坐标	焦距	高度/m
cam2513	(7037.1, 20.3), (-117.2, 126.1), (862.8, 24017.6)	2 015.5	8.1
cam9915	(-5026.3, -138.6), (722.7, 213.3), (-281.1, 13243.3)	1065.4	8.9

下载: 导出CSV

表 3 移动目标速度测量误差均值与方差

Table 3. Mean and variance of speed measurement errors for moving objects

视频名称	移动目标	特征点		边界框
视频名称	移动目标	均值/ (m· s^-1)	方差/ (m· s^-1)²	均值/ (m· s^-1)	方差/ (m· s^-1)²
cam2513-视频1	航空器	-1.1	0.3	13.8	600.4
cam2513-视频2	航空器	0	0	2.4	8.7
	航空器	-0.1	0.2	2.6	120.2
cam2513-视频3	车辆1	1.2	2.4	2.0	25.3
	车辆2	-0.8	0.1	-0.3	2.9
	车辆3	-1.9	0.3	-2.3	5.2
cam9915-视频1	航空器	-0.9	0.9	8.3	66.0
cam9915-视频1	车辆	-0.6	0	0.6	2.4
cam9915-视频2	车辆1	0.2	0.1	0.4	4.3
cam9915-视频2	车辆2	0.6	0.1	0.4	1.3
cam9915-视频3	航空器1	-0.1	0.2	0.6	2.9
	航空器2	-0.5	0.2	3.7	30.9
	车辆	-0.5	0.1	0.9	1.8

下载: 导出CSV

参考文献(19)

[1]	CERMENO E, PEREZ A, SIGUENZA J A.Intelligent video surveillance beyond robust background modeling[J].Expert Systems with Applications, 2018, 91:138-149.
[2]	罗晓, 卢宇, 吴宏刚.采用多视频融合的机场场面监视方法[J].电讯技术, 2011, 51(7):128-132. LUO X, LU Y, WU H G.A novel airport surface surveillance method using multi-video fusion[J].Telecommunication Engineering, 2011, 51(7):128-132(in Chinese).
[3]	VIDAKIS D G, KOSMOPOULOS D I.Facilitation of air traffic control via optical character recognition-based aircraft registration number extraction[J].IET Intelligent Transport Systems, 2018, 12(8):965-975.
[4]	LOPEZ-ARAQUISTAIN J, JARAMA A J, BESADA J A, et al.A new approach to map-assisted bayesian tracking filtering[J].Information Fusion, 2019, 45:79-95.
[5]	唐勇, 胡明华, 吴洪刚, 等.一种在机场视频中实现飞机自动挂标牌的新方法[J].江苏大学学报(自然科学版), 2013, 34(6):681-686. TANG Y, HU M H, WU H G, et al.An automatical labeling aircraft method for airport video monitoring[J].Journal of Jiangsu University(Natural Science Edition), 2013, 34(6):681-686(in Chinese).
[6]	SAIVADDI V, LU H L.Computer vision based surveillance concept for airport ramp operations[C]//Proceedings of 2013 IEEE/AIAA 32nd Digital Avionics Systems Conference(DASC).Piscataway, NJ: IEEE Press, 2013: 1-35.
[7]	LU H L, CHENG V H, TSAI J, et al.Airport gate operation monitoring using computer vision techniques[C]//Proceedings of 16th AIAA Aviation Technology, Integration, and Operations Conference.Reston: AIAA, 2016: 1-12.
[8]	DONADIO F, FREJAVILLE J, LARNIER S, et al.Artificial intelligence and collaborative robot to improve airport operations[C]//14th International Conference on Remote Engineering and Virtual Instrumentation(REV).Berlin: Springer, 2018: 973-986.
[9]	LU H L, KWAN J, FONG A, et al.Field testing of vision-based surveillance system for ramp area operations[C]//Proceedings of 2018 Aviation Technology, Integration, and Operations Conference.Reston: AIAA, 2018: 1-11.
[10]	CHEN J, WEISZER M, STEWART P, et al.Toward a more realistic, cost effective and greener ground movement through active routing:Part 1-Optimal speed profile generation[J].IEEE Transactions on Intelligent Transportation Systems, 2016, 17(5):1196-1209.
[11]	KANHERE N K, BIRCHFIELD S T.Real-time incremental segmentation and tracking of vehicles at low camera angles using stable features[J].IEEE Transactions on Intelligent Transportation Systems, 2008, 9(1):148-160.
[12]	詹昭焕, 韩松臣, 李炜, 等.基于倾向流和深度学习的机场运动目标检测[J].交通信息与安全, 2019, 37(1):49-57. ZHAN Z H, HAN S C, LI W, et al.A target detection method of moving objects at airport based on streak flow and deep learning[J].Journal of Transport Information and Safety, 2019, 37(1):49-57(in Chinese).
[13]	BARNICH O, VAN DROOGENBROECK M.ViBe:A universal background subtraction algorithm for video sequences[J].IEEE Transactions on Image Processing, 2011, 20(6):1709-1724.
[14]	ROSTEN E, PORTER R, DRUMMOND T.Faster and better:A machine learning approach to corner detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(1):105-119.
[15]	ILG E, MAYER N, SAIKIA T, et al.FlowNet 2.0: Evolution of optical flow estimation with deep networks[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Piscataway, NJ: IEEE Press, 2017: 1647-1655.
[16]	OCHS P, MALIK J, BROX T.Segmentation of moving objects by long term video analysis[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(6):1187-1200.
[17]	TOLDO R, FUSIELLO A.Robust multiple structures estimation with J-linkage[C]//Proceedings of European Conference on Computer Vision.Berlin: Springer, 2008: 537-547.
[18]	REDMON J, DIVVALA S, GIRSHICK R, et al.You only look once: Unified, real-time object detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Piscataway, NJ: IEEE Press, 2016: 779-788.
[19]	DANELLJAN M, BHAT G, KHAN F S, et al.ECO: Efficient convolution operators for tracking[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Piscataway, NJ: IEEE Press, 2017: 6638-6646.