شناسایی و ردیابی بی‌درنگ خودروها در سناریوهای حمل و نقل شهری براساس بینایی ماشین

نویسنده

دانشکده فنی و مهندسی، دانشگاه شاهد، ﺗﻬﺮان، اﻳﺮان

چکیده

در این مقاله یک روش قابل اعتماد شناسایی، کلاسه‌بندی و ردیابی خودروها در سناریوهای مختلف ترافیک درون شهری به‌ کمک تصاویر رنگی ویدیویی دریافتی از یک دوربین نظارتی ثابت تحت شرایط مختلف آب و هوایی ارایه می‌شود. این روش با استخراج خودکار تصویر زمینه و به‌روز کردن آن در چند فریم متوالی، مشکل تغییرات سریع و کند نور و محیط جاده را حل، و با موقعیت و مشکل کالیبراسیون دوربین تطبیق می‌یابد. ضمن عدم‌وابستگی به خطوط خط‌کشی و حرکت خودروها در بین خطوط، سایه‌های مزاحم خودی و پخشی خودرو، سایه‌های مزاحم کنار جاده و اثرات وزش باد، باران و برف را رفع می‌کند. همچنین برای تطبیق با تغییرات ساختاری و محیطی جاده، نیازی به حافظه یادگیری و پایگاه داده پیچیده ندارد. نتایج عملیاتی آزمایشات، مبین تشخیص دقیق خودروهای متحرک و عابرین، کارآیی بالا 99.8% در شناسایی و ردیابی، و 91% در کلاسه‌بندی خودرو است. با تنظیم فاصله و زاویه دوربین، شدت روشنایی و زمینه ساده تصویر دریافتی نتایج آزمایشات تجربی در سرعت و دقت تشخیص نتایج بهتری را نشان می‌دهد.

کلیدواژه‌ها

  • [1] Traffic Safety Facts 2003: A Compilation of MotorVehicle Crash Data from the Fatality Analysis ReportingSystem and the General Estimates System. US Departmentof Transportation, National Highway Traffic SafetyAdministration (NHTSA), Washington, D.C. 2004.
  • [2] A. Peters, S. von Klot, M. Heier, I. Trentinaglia, A.Hörmann, H. E. Wichmann, and H. Löwel, "Exposure toTraffic and the Onset ofMyocardial Infarction,"The NewEngland Journal of Medicine, vol. 351, no. 17, pp.1721-1730, 2004.
  • [3] J. J. Kim, S. Smorodinsky, M. Lipsett, B. C. Singer, A.T.Hodgson, and B. Ostro, "Traffic related Air Pollution nearBusy Roads: The East Bay Children’s Respiratory HealthStudy,"American Journal of Respiratory and Critical CareMedicine, vol. 170, pp. 520-526, 2004.
  • [4] A. Mahabadi, and A. Ranjbar, "Reliable License PlatRecognition,"The CSI journal on Computer Science andEngineering, vol. 7, no. 2&4 (b), pp. 51-67, 2009.
  • [5] G. Zhang, R. P. Avery, and Y. Wang, "A Video-basedVehicle Detection and Classification System for Real-timeTraffic Data Collection Using Uncalibrated Video Cameras,"Department of Civil and Environmental Engineering, pp. 2,2006.
  • [6] A. H. S. Lai, G. S. K. Fung, and N. H. C. Yung, "VehicleType Classification fromVisual-Based DimensionEstimation,"Proceedings of the IEEE IntelligentTransportation Systems Conference, Oakland, CA, pp.
  • 201-206, 2001.
  • [7] R. P. Avery, Y. Wang, and G. S. Rutherford, "Length- Based Vehicle Classification Using Images from th Uncalibrated Video Cameras,"Proceedings of the 7International IEEE Conferenceon Intelligent TransportationSystems, pp. 737-742, 2004.
  • [8] j. Bonneson, and M. Abbas, "Video Detection forIntersection and Interchange Control," FHWA/TX-03/4285-1, Texas Transportation Institute, College Station, Texas,2002.
  • [9] P. T. Martin, G. Dharmavaram, and A. Stevanovic, "Evaluation of UDOT’s Video Detection Systems: System’sPerformance in Various Test Conditions," Report No: UT-04.14. Salt Lake City, Utah, 2004.
  • [10] A. Rhodes, D. M. Bullock, J. Sturdevant, Z. Clark, andD. G. Candey, "Evaluation of Stop Bar Video DetectionAccuracy at Signalized Intersections," CD-ROM,Transportation Research Boardof the National Academies,Washington D.C., 2005.
  • [11] S. Gupte, O. Masoud, R. F. K. Martin, and N. P.Papanikolopoulos, "Detection and Classification ofVehicles,"IEEE Transactions on Intelligent TransportationSystems, vol. 3, no. 1, pp. 37-47, 2002.
  • [12] R. Cucchiara, C. Grana,M. Piccardi, and A. Prati, "Detecting Moving Objects, Ghosts and Shadows in VideoStreams,"IEEE Transactions on Pattern Analysis andMachine Intelligence, vol. 25, no. 10, pp. 1337-1342, 2003.
  • [13] N. Otsu, "A Threshold Selection Method from Gray-Level Histograms,"IEEE Transactions on Systems, Man andCybernetics, vol. 9, no. 1, pp. 62-66, 1979.
  • [14] R. P. Avery, G. Zhang, Y. Wang, and N. L. Nihan, "AnInvestigation into Shadow Removal from Traffic Images," 86th Annual Meeting of Transportation Research Board.Washington, D.C., 2006.
  • [15] L. G. Shapiro, and G. C. Stockman,Computer Vision,Prentice Hall, New Jersey: pp. 289-290, 2001.
  • [16] G. S. K. Fung, N. H. C. Yung, G. K. H. Pang, and A. H.S. Lai, "Effective Moving Cast Shadow Detection forMonocular Color Traffic Image Sequences,"OpticalEngineering, vol. 41, no. 6, pp. 1425-1440, 2002.
  • [17] J. M. Wang, Y. C. Chung, C. L. Chang, and S. W. Chen, "Shadow Detection and Removal for Traffic Images,"IEEEInternational Conference on Networking, Sensing andControl, vol. 1, pp. 649-654, 2004.
  • [18] J. Zheng, Y. Wang, N. L. Nihan, and E. Hallenbeck, "Extracting Roadway Background Image: a Mode-BasedApproach,"Journal ofTransportation Research Report, no.1994, pp. 82-88, 2006.
  • [19] Artificial Intelligence Laboratory at KyungpookNational University, Vehicle Detection & Tracking in aTraphicScene, http://ailib.knu.ac.kr/trafficsurveillancesystem/tracking.html,2015.
  • [20] D. Young, Sussex Computer Vision: Teach Vision 6,http://www.cogs.susx.ac.uk/users/davidy/teachvision/vision6 .html, 2015.
  • [21] D. Beymer, P. McLauchlan, B. Coifman, and J. Malik, "A Real-time Computer VisionSystem for Measuring TrafficParameters,"Proceedings of the IEEE Computer SocietyConference on Computer vision and Pattern Recognition (CVPR), pp. 495-501, 1997.
  • [23] A. Mahabadi, "Intelligent Traffic Light,"The CSIjournal on Computer Science and Engineering, vol. 7, no. 2,pp. 51-67, 2009.
  • [24] A. B. Hillel, R. Lerner, D. Levi, and G. Raz, "Recentprogress in road and lane detection: a survey,"MachineVision and Applications, no. 25, pp. 727–745, 2014.
  • [25] T. Gao, P. Wang, C. Wang, Z. Yao, "Feature ParticlesTracking for Moving Objects,"JOURNAL OFMULTIMEDIA, vol. 7, no. 6, pp. 408-414, 2012.
  • [26] C. Chi, M. Ku, and C. Wang, "Automatic TrafficSurveillance System for Vision-Based Vehicle Recognitionand Tracking,"Journal of Information Science andEngineering, no. 26, pp. 611-629, 2010.
  • [27] D. Devarajan, Z. Cheng, and R. Radke, "Calibratingdistributed camera networks,"IEEE proceeding,vol. 96, no. 10, OCT, pp. 1625–1639, 2008.
  • [28] M. Darms, P. Rybski, and C. Urmson, Vehicle detectionand tracking for the urban challenge, inAutomatisierungs,Assistenz und Eingebettete Systeme für TransportmittelSymp.,Braunschweig, Germany, 2008.
  • [29] Y. Li, B. Li, and D.Wen, "Rear-View VehicleDetection and Tracking by Combining Multiple Parts forComplex Urban Surveillance," Intelligent TransportationSystems, IEEE Transactions,vol. 15, no. 2, pp. 597-606,2014.
  • [30] V. Ramakrishnan, A. K. Prabhavathy, and J. Devishree, "A Survey on Vehicle Detection Techniques in AerialSurveillance," International Journal of ComputerApplications, vol. 55, no.18, pp. 43-47, 2012.
  • [31] H. C. Choi, J. M. Park, W. S. Choi, and S. Y. Oh, "Vision-based Fusion of Robust Lane Tracking and ForwardVehicle Detection in a Real Driving Environment,"
  • International Journal of Automotive Technology, vol. 13, no.4, pp. 653– 669, 2012.
  • [32]H. Kong,J. Audibert,and J. Ponce, "Generalroaddetection from asingle image,"IEEE Transaction on ImageProcessing, vol.19, no. 8, pp.2211–2220, 2010.
  • [33]Y. Li, B.Li, and D.Wen, "Rear-View VehicleDetection and Tracking byCombiningMultiple PartsforComplex UrbanSurveillance,"IEEETransactionon Intelligent Transportation Systems, vol. 15, Issue. 2,pp.597–606, 2014.
  • [34]C. Rasmussen, and T.Korah, "On-vehicle and aerialtexture analysisfor vision-based desertroad following," CVPR Workshop on Machine Visionfor IntelligentVehicles,vol. III,pp. 66-66, 2005.
  • [35]S. Kammel,and B. Pitzer, "LIDAR-based lane markerdetection andMapping,"IEEE Intelligent VehiclesSymposium, pp. 1137–1142, 2008.
  • [36]A. S. Huang, D. Moore,M. Antone,E. Olson, andS.Teller, "Findingmultiple lanesin urban road networkswithvision and LIDAR,"Auton.Robots, no.26, pp.103–122,
  • 2009
  • [37]J. McCall, and M. Trivedi, "Video-based lane estimationandtracking fordriver assistance: survey,"systemandevaluation, IEEETransaction IntellgentTransportationSystem, no. 7, pp.20–37, 2006.
  • [38]Y. Yi,Onimproving the accuarcyand reliabilityofGPS/INS baseddirect sensor georeferencing, Ph.D.Dissertation, OhioState University, Columbus, 2007.
  • [39]A. B, Hillel,R. Lerner,D. Levi, andG. Raz , "Recentprogress in roadand lane detection: a survey,"MachineVision and Applications, no. 25, pp. 727–745, 2014.
  • [40]C. J. V.Rijsbergen,Information retrieval, London:Butterworth & CoLtd (Publishers), secondedition, 1979.
دوره 13، شماره 1
بهار و تابستان
اردیبهشت 1394