Skip to main content
SpringerLink
Log in

A Two-Stage Framework for Diverse Traffic Light Recognition Based on Individual Signal Detection

  • Conference paper
  • First Online:
Pattern Recognition and Artificial Intelligence (MedPRAI 2021)

Part of the book series: Communications in Computer and Information Science ((CCIS,volume 1543))

Abstract

The detection of traffic lights is an indispensable part of advanced driver assistance systems. This paper presents a new two-stage detection framework which is able to recognize various types of traffic signals. Different from the techniques based on the direct detection of traffic lights, we consider the individual signal bulbs as targets for detection and recognition. In our two-stage approach, the first detection stage aims to achieve a very low miss rate on the traffic signal bulbs with possibly high false positives. It is then followed by the second recognition stage to single out the correct traffic signals using a classification network. The proposed method overcomes the diverse traffic light detection problem due to various arrangements of signal bulbs in different countries. It is also capable of simultaneously detecting individual traffic signals including the arrow lights. Thus, the categories of traffic light states for classification and training can be reduced. The experiments carried out using our road scene images and public datasets have demonstrated the feasibility of the proposed technique.

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

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 69.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 89.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    https://www.acelab.org/traffic-light-dataset.

  2. 2.

    https://www.mapillary.com/.

References

  1. Bach, M., Stumper, D., Dietmayer, K.: Deep convolutional traffic light recognition for automated driving. In: 2018 21st International Conference on Intelligent Transportation Systems (ITSC), pp. 851–858. IEEE (2018)

    Google Scholar 

  2. Behrendt, K., Novak, L., Botros, R.: A deep learning approach to traffic lights: detection, tracking, and classification. In: 2017 IEEE International Conference on Robotics and Automation (ICRA), pp. 1370–1377 (2017). https://doi.org/10.1109/ICRA.2017.7989163

  3. Behrendt, K., Novak, L., Botros, R.: A deep learning approach to traffic lights: detection, tracking, and classification. In: 2017 IEEE International Conference on Robotics and Automation (ICRA), pp. 1370–1377. IEEE (2017)

    Google Scholar 

  4. Bochkovskiy, A., Wang, C.Y., Liao, H.Y.M.: Yolov4: Optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934 (2020)

  5. Chaudhury, K., DiVerdi, S., Ioffe, S.: Auto-rectification of user photos. In: 2014 IEEE International Conference on Image Processing (ICIP), pp. 3479–3483 (2014). https://doi.org/10.1109/ICIP.2014.7025706

  6. Chen, Z., Huang, X.: Accurate and reliable detection of traffic lights using multiclass learning and multiobject tracking. IEEE Intell. Transp. Syst. Mag. 8(4), 28–42 (2016). https://doi.org/10.1109/MITS.2016.2605381

    Article  Google Scholar 

  7. Cordts, M., et al.: The cityscapes dataset for semantic urban scene understanding. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)

    Google Scholar 

  8. Fregin, A., Muller, J., Krebel, U., Dietmayer, K.: The driveu traffic light dataset: introduction and comparison with existing datasets. In: 2018 IEEE International Conference on Robotics and Automation (ICRA), pp. 3376–3383 (2018). https://doi.org/10.1109/ICRA.2018.8460737

  9. Girshick, R.: Fast r-cnn. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1440–1448 (2015)

    Google Scholar 

  10. Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 580–587 (2014)

    Google Scholar 

  11. Jensen, M.B., Philipsen, M.P., Møgelmose, A., Moeslund, T.B., Trivedi, M.M.: Vision for looking at traffic lights: issues, survey, and perspectives. IEEE Trans. Intell. Transp. Syst. 17(7), 1800–1815 (2016). https://doi.org/10.1109/TITS.2015.2509509

    Article  Google Scholar 

  12. John, V., Zheming, L., Mita, S.: Robust traffic light and arrow detection using optimal camera parameters and gps-based priors. In: 2016 Asia-Pacific Conference on Intelligent Robot Systems (ACIRS), pp. 204–208. IEEE (2016)

    Google Scholar 

  13. Lin, T., Dollár, P., Girshick, R., He, K., Hariharan, B., Belongie, S.: Feature pyramid networks for object detection. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 936–944 (2017). https://doi.org/10.1109/CVPR.2017.106

  14. Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: unified, real-time object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 779–788 (2016)

    Google Scholar 

  15. Redmon, J., Farhadi, A.: Yolo9000: better, faster, stronger. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 7263–7271 (2017)

    Google Scholar 

  16. Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: towards real-time object detection with region proposal networks. In: Advances in Neural Information Processing Systems, pp. 91–99 (2015)

    Google Scholar 

  17. Weber, M., Huber, M., Zöllner, J.M.: Hdtlr: a CNN based hierarchical detector for traffic lights. In: 2018 21st International Conference on Intelligent Transportation Systems (ITSC), pp. 255–260. IEEE (2018)

    Google Scholar 

  18. Weber, M., Wolf, P., Zöllner, J.M.: Deeptlr: a single deep convolutional network for detection and classification of traffic lights. In: 2016 IEEE Intelligent Vehicles Symposium (IV), pp. 342–348. IEEE (2016)

    Google Scholar 

  19. Yeh, T.W., Lin, H.Y.: Detection and recognition of arrow traffic signals using a two-stage neural network structure. In: The 6th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS) (2020)

    Google Scholar 

  20. Yeh, T.W., Lin, S.Y., Lin, H.Y., Chan, S.W., Lin, C.T., Lin, Y.Y.: Traffic light detection using convolutional neural networks and lidar data. In: 2019 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS), pp. 1–2. IEEE (2019)

    Google Scholar 

  21. Zhao, H., Qi, X., Shen, X., Shi, J., Jia, J.: ICNet for real-time semantic segmentation on high-resolution images. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11207, pp. 418–434. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01219-9_25

    Chapter  Google Scholar 

Download references

Acknowledgment

The support of this work in part by the Ministry of Science and Technology of Taiwan under Grant MOST 106-2221-E-194-004 and the Advanced Institute of Manufacturing with High-tech Innovations (AIM-HI) from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan is gratefully acknowledged.

Author information

Authors and Affiliations

  1. Department of Electrical Engineering, National Chung Cheng University, Chiayi 621, Taiwan

    Ssu-Yun Lin & Huei-Yung Lin

Authors
  1. Ssu-Yun Lin
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Huei-Yung Lin
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Huei-Yung Lin .

Editor information

Editors and Affiliations

  1. Larbi Tebessi University, Tebessa, Algeria

    Chawki Djeddi

  2. Bahria University, Islamabad, Pakistan

    Imran Siddiqi

  3. FAST National University, Islamabad, Pakistan

    Akhtar Jamil

  4. Istanbul Sabahattin Zaim University, Istanbul, Turkey

    Alaa Ali Hameed

  5. Istanbul Sabahattin Zaim University, Istanbul, Turkey

    İsmail Kucuk

Rights and permissions

Reprints and permissions

Copyright information

© 2022 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Lin, SY., Lin, HY. (2022). A Two-Stage Framework for Diverse Traffic Light Recognition Based on Individual Signal Detection. In: Djeddi, C., Siddiqi, I., Jamil, A., Ali Hameed, A., Kucuk, İ. (eds) Pattern Recognition and Artificial Intelligence. MedPRAI 2021. Communications in Computer and Information Science, vol 1543. Springer, Cham. https://doi.org/10.1007/978-3-031-04112-9_20

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-04112-9_20

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-04111-2

  • Online ISBN: 978-3-031-04112-9

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation