Skip to main content
SpringerLink
Log in

A deep learning-assisted mathematical model for decongestion time prediction at railroad grade crossings

  • Original Article
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

A Correction to this article was published on 28 November 2021

This article has been updated

Abstract

This paper presents a deep learning-assisted framework to estimate the decongestion time at the grade crossing, and its key novelty lies in a differential approach to address the challenge associated with data deficiency of congestion events in grade crossings. A hypothesis of the traffic behavior during the congestion event caused by passing trains is proposed. A deep neural network-based vehicle crowd counting algorithm is developed to estimate the number of vehicles at the normal traffic condition. A running average-based motion detection algorithm is designed to estimate the time of the train passing through the grade crossing. A regression model is then constructed to relate the quantitative information with the decongestion time. In the experiments, 30 congestion events are video-recorded during a period of 200 h with different camera angles at a selected grade crossing, and then studied by the proposed method to learn the congestion pattern and predict the decongestion time, which to the best of our knowledge has not been attempted before. Analysis of the experimental results shows that the vehicle number at the normal traffic flow and the train passing time have significant influences on the traffic decongestion time. The relationship is captured by a quantitative model for rapid prediction. Our study also points out the direction for further improvement of the present development to meet the need for real-world applications.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14

Similar content being viewed by others

Change history

Abbreviations

FRA:

Federal Railroad Administration

DNN:

Deep Neural Network

ROI:

Region of Interest

ETT:

Estimated Total Time

ETP:

Estimated Time of Train Passing

ETD:

Estimated Time of Decongestion

NTC:

Normal Traffic Condition

GT:

Ground Truth

NNTC:

Non-Normal Traffic Condition

GT-NNTC:

Ground Truth of Non-Normal Traffic Condition

GT-NTC:

Ground Truth of Normal Traffic Condition

GT-ETP:

Ground Truth of Estimated Time of Train Passing

GT-ETD:

Ground Truth of Estimated Time of Decongestion

CNN:

Convolutional Neural Network

MAE:

Mean Absolute Error

References

  1. Alexy K (2020) Tracking toward zero: improving grade crossing safety and addressing community concerns. https://www.transportation.gov/print/364121

  2. Arnott R, Palma A, Lindsey R (1991) Does providing information to drivers reduce traffic congestion? Trans Research Part A General 25(5):309–318

    Article  Google Scholar 

  3. Bando M, Hasebe K, Nakayama A, Shibata A, Sugiyama Y (1995) Dynamical model of traffic congestion and numerical simulation. Phys Rev E 51(2):1035–1042

    Article  Google Scholar 

  4. Bauza R, Gozalvez J (2013) Traffic congestion detection in large-scale scenarios using vehicle-to-vehicle communications. Jour Netw Comput Appl 36(5):1295–1307

    Article  Google Scholar 

  5. Chan AB, Vasconcelos N (2009) Bayesian Poisson regression for crowd counting. In: Proc IEEE Conf Comp Vis (ICCV), Kyoto

  6. Chan AB, Liang ZJ, Vasconcelos N (2008) Privacy preserving crowd monitoring: Counting people without people models or tracking. In: Proc. IEEE Conf. Comp. Vis. Pattern Recognit. (CVPR), Anchorage

  7. Chu T, Wang J, Codeca L, Li Z (2019) Multi-agent deep reinforcement learning for large-scale traffic signal control. arXiv preprint arXiv:1903.04527

  8. Fan Q, Zhuo W, Tang C, Tai Y (2020) Few-shot object detection with attention-rpn and multi-relation detector. In: Proc IEEE Conf Comp Vis Pattern Recognit (CVPR), Online conference

  9. Federal Railroad Administration (2019) Highway-rail crossing handbook - Third Edition. https://safety.fhwa.dot.gov/hsip/xings/com_roaduser/fhwasa18040/fhwasa18040v2.pdf

  10. Ge W, Collins RT (2009) Marked point processes for crowd counting. In: Proc IEEE Conf Comp Vis Pattern Recognit (CVPR), Miami

  11. Gephardt M, Poe M (2018) Trains blocking roadway creates dangerous situations, first responders say. https://kutv.com/news/get-gephardt/trains

  12. Gruden C, Otkovic II, Sraml M (2020) Neural networks applied to microsimulation: A prediction model for pedestrian crossing time. Sustainability 12(13):1–22

    Article  Google Scholar 

  13. Hayda A, Yilmaz Y (2018) An efficient deep reinforcement learning model for urban traffic control. arXiv preprint arXiv:2005.00935v1

  14. Hayda A, Yilmaz Y (2020) Deep reinforcement learning for intelligent transportation systems: a survey. arXiv preprint arXiv:2005.00935v1

  15. Hensher DA (1994) Hazard-based duration models and their application to transport analysis. Trans Rev 14(1):63–82

    Article  Google Scholar 

  16. Hoel CJ, Driggs-Campbell K, Wolff K, Laine L, Kochenderfer MJ (2019) Combining planning and deep reinforcement learning in tactical decision making for autonomous driving. arXiv preprint arXiv:1905.02680

  17. Huang W, Song G, Hong H, Xie K (2014) Deep architecture for traffic flow prediction: deep belief networks with multitask learning. IEEE Trans Intell Trans Syst 15(5):2191–2201

    Article  Google Scholar 

  18. Jiang X, Xiao Z, Zhang B, Zhen X, Cao X, Doermann D, Shao L (2019) Crowd counting and density estimation by trellis encoder-decoder networks. In: Proc IEEE Conf Comp Vis Pattern Recognit (CVPR), Long Beach

  19. Jiang Z (2021) Github codes. https://github.com/jj574435561/Decongestion-time-prediction-at-Railroad-Grade-Crossing

  20. Jin J, Ma X (2019) A multi-objective agent-based control approach with application in intelligent traffic signal system. IEEE Trans Intell Trans Syst 20(10):3900–3912

    Article  Google Scholar 

  21. Ke W, Zhang T, Huang Z, Ye Q, Liu J, Huang D (2020) Multiple anchor learning for visual object detection. In: Proc IEEE Conf Comp Vis Pattern Recognit (CVPR), Online conference

  22. Khan AM (2010) Prediction and display of delay at road border crossings. The Open Trans J 4:9–22

    Article  Google Scholar 

  23. Komatsu TS, Sasa S (1995) Kink soliton characterizing traffic congestion. Phys Rev E 52(5):5574–5582

    Article  Google Scholar 

  24. Kurniawan J, Syahra SG, Dewa CK, Afiahayati, (2018) Traffic congestion detection: learning from CCTV monitoring images using convolutional neural network. Procedia Comput Sci 144:291–297

    Article  Google Scholar 

  25. Lee K, Hong B, Jeong D, Lee J (2014) Congestion pattern model for predicting short-term traffic decongestion times. In: Proc IEEE Conf Intell Transp Sys, Qinghao

  26. Lempitsky V, Zisserman A (2010) Learning to count objects in images. In: Proc Neural Inf Process Sys (NeurIPS), Vancouver

  27. Li M, Zhang Z, Huang K, Tan T (2008) Estimating the number of people in crowded scenes by mid based foreground segmentation and head-shoulder detection. In: Proc Int Conf Pattern Recognit, Tampa

  28. Li Y, Zhang X, Chen D (2018) CSRNet: Dilated convolutional neural networks for understanding the highly congested scenes. arXiv preprint arXiv:1802.10062

  29. Liu C, Weng X, Mu Y (2019a) Recurrent attentive zooming for joint crowd counting and precise localization. In: Proc IEEE Conf Comp Vis Pattern Recognit. (CVPR), Long Beach

  30. Liu W, Salzmann M, Fua P (2019b) Context-aware crowd counting. In: Proc. IEEE Conf. Comp. Vis. Pattern Recognit. (CVPR), Long Beach

  31. Lv Y, Duan Y, Kang W, Li Z, Wang FY (2015) Traffic flow prediction with big data: a deep learning approach. IEEE Trans Intell Trans Syst 16(2):865–873

    Google Scholar 

  32. Ma X, Yu H, Wang Y, Wang Y (2015) Large-scale transportation network congestion evolution prediction using deep learning theory. PLOS ONE 10(3):1–17

    Google Scholar 

  33. Njus E (2016) How long can trains block railroad crossings? https://www.oregonlive.com/commuting/2016/11/how_long_can_trains_block_rail.html

  34. Paselk TA, Mannering FL (1994) Use of duration models for predicting vehicular delay at a US/Canadian border crossing. Transportation 21:865–873

    Article  Google Scholar 

  35. Polson NG, Sokolov VO (2017) Deep learning for short-term traffic flow prediction. Transp Res Part C Emerg Technol 79:1–17

    Article  Google Scholar 

  36. Ranjan V, Le H, Hoai M (2018) Iterative crowd counting. In: Proc Eur Conf Comp Vis (ECCV), Munich

  37. Sallab AE, Abdou M, Perot E, Yogamani S (2017) Deep reinforcement learning framework for autonomous driving. arXiv preprint arXiv:1704.02532

  38. Sam DB, Surya S, Babu RV (2017) Switching convolutional neural network for crowd counting. In: Proc IEEE Conf Comp Vis Pattern Recognit (CVPR), Hawaii

  39. Sam DB, Sajjan NN, Babu RV, Srinivasan M (2018) Divide and grow: Capturing huge diversity in crowd images with incrementally growing CNN. In: Proc IEEE Conf Comp Vis Pattern Recognit (CVPR), Salt Lake City

  40. Shi Z, Zhang L, Liu Y, Cao X, Ye Y, Cheng M, Zheng G (2018) Crowd counting with deep negative correlation learning. In: Proc IEEE Conf Comp Vis Pattern Recognit (CVPR), Salt Lake City

  41. Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556

  42. Tan M, Pang R, Le QV (2020) EfficientDet: Scalable and efficient object detection. In: Proc IEEE Conf Comp Vis Pattern Recognit (CVPR), Online conference

  43. Veres M, Moussa M (2020) Deep learning for intelligent transportation systems: a survey of emerging trends. IEEE Trans Intell Trans Syst 21(8):3152–3168

    Article  Google Scholar 

  44. Walach E, Wolf L (2016) Learning to count with CNN boosting. In: Proc Eur Conf Comp Vis (ECCV), Amsterdam

  45. Wan J, Yuan Y, Wang Q (2017) Traffic congestion analysis: a new perspective. In: Proc IEEE Int Conf Acoustics, speech and Signal Processing (ICASSP), New Orleans

  46. Wang J, Gu Q, Wu J, Liu G, Xiong Z (2016) Traffic speed prediction and congestion source exploration: a deep learning method. In: Proc IEEE Int Conf Data Mining (ICDM), Barcelona

  47. Wang Q, Wan J, Yuan Y (2018) Locality constraint distance metric learning for traffic congestion detection. Pattern Recognit 75:272–281

    Article  Google Scholar 

  48. Wu C, Kreidieh A, Parvate K, Vinitsky E, Bayen AM (2017) Flow: a modular learning framework for autonomy in traffic. arXiv preprint arXiv:1710.05465

  49. Xiong F, Shi X, Yeung DY (2017) Spatiotemporal modeling for crowd counting in videos. In: Proc IEEE Conf Comp Vis (ICCV), Venice

  50. Xiong X, Wang J, Zhang F, Li K (2016) Combining deep reinforcement learning and safety based control for autonomous driving. arXiv preprint arXiv:1612.00147

  51. Yau KA, Qadir J, Kho HL, Ling MH, Komisarczuk P (2017) A survey on reinforcement learning models and algorithms for traffic signal control. ACM Comput Surv 50(34)

  52. Zhang S, Yao Y, Hu J, Zhao Y, Li S, Hu J (2019) Deep autoencoder neural networks for short-term traffic congestion prediction of transportation networks. Sensors (Basel) 19(10):1–19

    Google Scholar 

  53. Zhao Z, Chen W, Wu X, Chen PC, Liu J (2017) LSTM network: a deep learning approach for short-term traffic forecast. IET Intell Trans Syst 11(2):68–75

    Article  Google Scholar 

Download references

Acknowledgements

This research is partially funded by the Federal Railroad Administration (FRA), Contract No. 693JJ6-19-C-000009. Mr. Francesco Bedini, Dr. Shala Blue, and Dr. Starr Kidda from FRA have provided important guidance and insight. Special thanks to Mayor Steve Benjamin, the City Council, and the first responders at the City of Columbia for their tremendous support. Mr. Robert Anderson, Mr. David Brewer, Mr. John Stroke and other members at the Public Works Department provided great help for the traffic monitoring. The opinions expressed in this article are solely those of the authors and do not represent the opinions of the funding agencies.

Author information

Authors and Affiliations

  1. Department of Mechanical Engineering, University of South Carolina, Columbina, SC, 29208, USA

    Zhuocheng Jiang & Yi Wang

  2. Department of Civil and Environment Engineering, University of South Carolina, Columbina, SC, 29208, USA

    Feng Guo & Yu Qian

  3. Department of Electrical and Computer Engineering, University of Alabama in Huntsville, Huntsville, AL, 35899, USA

    W. David Pan

Authors
  1. Zhuocheng Jiang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Feng Guo
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yu Qian
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yi Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. W. David Pan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Yi Wang.

Ethics declarations

Conflicts of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jiang, Z., Guo, F., Qian, Y. et al. A deep learning-assisted mathematical model for decongestion time prediction at railroad grade crossings. Neural Comput & Applic 34, 4715–4732 (2022). https://doi.org/10.1007/s00521-021-06625-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-021-06625-z

Keywords

Search

Navigation