Abstract
Vehicle trajectories provide very useful empirical data for studying traffic phenomena such as vehicle following behavior, lane changing behavior, traffic oscillations, capacity drop, safety analysis, etc. However, there are a very limited number of studies on extracting trajectory data from mixed traffic and for congested conditions. This paper presents a deep learning-based framework to extract vehicle trajectories in mixed traffic under both free-flow and congested conditions. The popular YOLOv3 deep learning architecture is used and trained on a hybrid dataset generated from two different sets of frames with different scales and orientations. The anchor boxes for vehicle detection and classification are customized to improve accuracy and efficiency. The SORT algorithm is used to track the identified vehicles and the extracted trajectory data are benchmarked with a popular trajectory extraction portal that showed that the proposed model performs well for trajectory extraction. The paper also presents a methodology based on numerical integration techniques to impute missing trajectory data. Finally, the trajectory data obtained from the adjacent road sections are aligned and scaled to the real-world coordinates using coordination transformation and error correction methods to make it useful for research purposes. The extracted trajectories show remarkable accuracy with approximately 0.25–0.35 m of precision. It is expected that these trajectories capture traffic and driving behavior phenomena for a better understanding of mixed traffic conditions.
Similar content being viewed by others
References
Adu-Gyamfi YO, Asare SK, Sharma A, Titus T (2017) Automated vehicle recognition with deep convolutional neural networks. Transp Res Rec 2645(1):113–122
Apeltauer J, Babinec A, Herman D, Apeltauer T (2015) Automatic vehicle trajectory extraction for traffic analysis from aerial video data. Int Arch Photogramm Remote Sensing Spatial Inform Sci 40(3):9
Barmpounakis EN, Vlahogianni EI, Golias JC (2016) Extracting kinematic characteristics from unmanned aerial vehicles (No. 16–3429)
Bay H, Ess A, Tuytelaars T, Van Gool L (2008) Speeded-up robust features (SURF). Comput Vis Image Underst 110(3):346–359
Bewley A, Ge Z, Ott L, Ramos F, Upcroft B (2016) Simple online and realtime tracking. In 2016 IEEE International Conference on Image Processing (ICIP) 3464–3468
Bisong E (2019) Google Colaboratory. In: Building Machine Learning and Deep Learning Models on Google Cloud Platform. A press, Berkeley https://doi.org/10.1007/978-1-4842-4470-8_7
Chen E, Tang X, Fu B (2018) A modified pedestrian retrieval method based on faster R-CNN with integration of pedestrian detection and re-identification. In: 2018 International conference on audio, language and image processing (ICALIP) pp 63–66
Chen X, Li Z, Yang Y, Qi L, Ke R (2020) High-resolution vehicle trajectory extraction and denoising from aerial videos. IEEE Transactions on Intelligent Transportation Systems
Cheng HY, Yu CC (2012) Detecting and Tracking Vehicles in Airborne Videos. International Journal of Computer and Information Engineering 6(5):665–668
Choi JH, Lee D, Bang H (2011) Tracking an unknown moving target from uav: Extracting and localizing an moving target with vision sensor based on optical flow. In The 5th International Conference on Automation, Robotics and Applications 384–389
Coifman B, Li L (2017) A critical evaluation of the Next Generation Simulation (NGSIM) vehicle trajectory dataset. Transportation Research Part B: Methodological 105:362–377
Danescu R, Oniga F, Nedevschi S, Meinecke MM (2009) Tracking multiple objects using particle filters and digital elevation maps. In 2009 IEEE Intelligent Vehicles Symposium 88–93
DFS (Data From Sky, 2020): https://datafromsky.com/
Gao H, Kong SL, Zhou S, Lv F, Chen Q (2014) Automatic extraction of multi-vehicle trajectory based on traffic videotaping from quadcopter model. Appl Mechan Mater 552:232–239. Trans Tech Publications Ltd.
Garcia F, Cerri P, Broggi A, de la Escalera A, Armingol JM (2012) Data fusion for overtaking vehicle detection based on radar and optical flow. In 2012 IEEE Intelligent Vehicles Symposium 494–499
Gould, H., Tobochnik, J., & Christian, W. (2016). An introduction to computer simulation methods application to physical system.
Guido G, Vitale A, Saccomanno FF, Astarita V, Giofrè V (2014) Vehicle tracking system based on videotaping data. Procedia Soc Behav Sci 111:1123–1132
Haghighat AK, Ravichandra-Mouli V, Chakraborty P, Esfandiari Y, Arabi S, Sharma A (2020) Applications of Deep Learning in Intelligent Transportation Systems. Journal of Big Data Analytics in Transportation, 2(2):115–145
He K, Gkioxari G, Dollár P, Girshick RB (2017) "Mask R-CNN", 2017 IEEE International Conference on Computer Vision (ICCV), pp. 2980–2988
Henriques JF, Caseiro R, Martins P, Batista J (2014) High-speed tracking with kernelized correlation filters. IEEE Trans Pattern Anal Mach Intell 37(3):583–596
Hue C, Le Cadre JP, Pérez P (2002) Sequential Monte Carlo methods for multiple target tracking and data fusion. IEEE Trans Signal Process 50(2):309–325
Jonathan Hui, “Object detection: speed and accuracy comparison (Faster R-CNN, R-FCN, SSD, FPN, RetinaNet and YOLOv3)”, https://medium.com/@jonathan_hui/object-detection-speed-and-accuracy-comparison-faster-r-cnn-r-fcn-ssd-and-yolo-5425656ae359
Jodoin JP, Bilodeau GA, Saunier N (2014) Urban tracker: Multiple object tracking in urban mixed traffic. In IEEE Winter Conference on Applications of Computer Vision 885–892
Kanagaraj V, Asaithambi G, Toledo T, Lee TC (2015) Trajectory data and flow characteristics of mixed traffic. Transp Rese Rec 2491(1):1–11
Khan MA, Ectors W, Bellemans T, Janssens D, Wets G (2017) Unmanned aerial vehicle–based traffic analysis: Methodological framework for automated multivehicle trajectory extraction. Transp Res Rec 2626(1):25–33
Kim Z, Gomes G, Hranac R, Skabardonis A (2005) A machine vision system for generating vehicle trajectories over extended freeway segments. In 12th World Congress on Intelligent Transportation Systems
Kim EJ, Park HC, Ham SW, Kho SY, Kim DK (2019) Extracting vehicle trajectories using unmanned aerial vehicles in congested traffic conditions. J Adv Transp. https://doi.org/10.1155/2019/9060797
Koller D, Weber J, Malik J (1994) Robust multiple car tracking with occlusion reasoning. European Conference on Computer Vision. Springer, Berlin, pp 189–196
Leibe B, Seemann E, Schiele B (2005) Pedestrian detection in crowded scenes. In 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) 1:878–885
Li Q, Lei B, Yu Y, Hou R (2009) Real-time highway traffic information extraction based on airborne video. In 2009 12th International IEEE Conference on Intelligent Transportation Systems. IEEE. 1–6
Li J, Hajimirsadeghi H, Zaki MH, Mori G, Sayed T (2014) Computer vision techniques to collect helmet-wearing data on cyclists. Transp Res Rec 2468(1):1–10
Miao Q, Wang G, Shi C, Lin X, Ruan Z (2011) A new framework for on-line object tracking based on SURF. Pattern Recogn Lett 32(13):1564–1571
Milan A, Roth S, Schindler K (2013) Continuous energy minimization for multitarget tracking. IEEE Trans Pattern Anal Mach Intell 36(1):58–72
Montanino M, Punzo V (2013) Making NGSIM data usable for studies on traffic flow theory: Multistep method for vehicle trajectory reconstruction. Transp Res Rec 2390(1):99–111
Oh J, Min J, Kim M, Cho H (2009) Development of an automatic traffic conflict detection system based on image tracking technology. Transp Res Rec 2129(1):45–54
Ozkurt C, Camci F (2009) Automatic traffic density estimation and vehicle classification for traffic surveillance systems using neural networks. Mathem Comput Appl 14(3):187–196
Redmon, Joseph, et al (2016) "You only look once: Unified, real-time object detection." Proceedings of the IEEE conference on computer vision and pattern recognition
Redmon J, Farhadi A (2018) Yolov3: An incremental improvement. arXiv preprint
Ren S, He K, Girshick R Sun J (2015) Faster r-cnn: Towards real-time object detection with region proposal networks. Adv Neural Inform Process Syst 91–99
Rodríguez-Canosa GR, Thomas S, Del Cerro J, Barrientos A, MacDonald B (2012) A real-time method to detect and track moving objects (DATMO) from unmanned aerial vehicles (UAVs) using a single camera. Remote Sensing 4(4):1090–1111
Rothrock RL, Drummond OE (2000) Performance metrics for multiple-sensor multiple-target tracking. Signal Data Process Small Targets 4048:521–531
Song H, Liang H, Li H, Dai Z, Yun X (2019) Vision-based vehicle detection and counting system using deep learning in highway scenes. European Transport Research Review, 11(1):1–16
St-Aubin P, Saunier N, Miranda-Moreno LF, Ismail K (2013) Use of computer vision data for detailed driver behavior analysis and trajectory interpretation at roundabouts. Transp Res Rec 2389(1):65–77
Tsai LW, Hsieh JW, Fan KC (2007) Vehicle detection using normalized color and edge map. IEEE Trans Image Process 16(3):850–864
Tzutalin (2015) LabelImg. Git code. https://github.com/tzutalin/labelImg
Wang L, Yung NHC (2012) Three-dimensional model-based human detection in crowded scenes. IEEE Trans Intell Transp Syst 13(2):691–703
Wang L, Yung NHC, Xu L (2014) Multiple-human tracking by iterative data association and detection update. IEEE Trans Intell Transp Syst 15(5):1886–1899
Wang L, Chen F, Yin H (2016). Detecting and tracking vehicles in traffic by unmanned aerial vehicles. Automation in construction, 72:294–308
Xu Y, Yu G, Wang Y, Wu X, Ma Y (2017) Car detection from low-altitude UAV imagery with the faster R-CNN. J Adv Transp. https://doi.org/10.1155/2017/2823617
Yilmaz A, Javed O, Shah M (2006) Object tracking: A survey. Acm Comput Surveys (CSUR) 38(4):13-es
Acknowledgements
The authors would also like to thank Kiran Roy, a former Dual Degree student in the Department of Civil Engineering, IIT Madras for recording the videos, IIT Madras for providing the facilities to conduct the research, and MHRD, Government of India for providing the scholarship to the first author.
Funding
The authors would also like to thank the SPARC program, MHRD, Government of India, and IC&SR, IITM for supporting this research through projects.
Author information
Authors and Affiliations
Contributions
Conceptualization: Bhargava Rama Chilukuri; Methodology: Rohan Dhatbale, Bhargava Rama Chilukuri; Formal analysis and investigation: Rohan Dhatbale, Bhargava Rama Chilukuri; Writing—original draft preparation: Rohan Dhatbale; Writing—review and editing: Rohan Dhatbale, Bhargava Rama Chilukuri; Funding acquisition: Bhargava Rama Chilukuri; Resources: IIT Madras, MHRD (Govt. of India); Supervision: Bhargava Rama Chilukuri.
Corresponding author
Ethics declarations
Conflicts of interest
None.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Dhatbale, R., Chilukuri, B.R. Deep Learning Techniques for Vehicle Trajectory Extraction in Mixed Traffic. J. Big Data Anal. Transp. 3, 141–157 (2021). https://doi.org/10.1007/s42421-021-00042-3
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s42421-021-00042-3