Deep Learning for Detection of Railway Signs and Signals

  • Georgios KaragiannisEmail author
  • Søren Olsen
  • Kim Pedersen
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 943)


Major railway lines need advance management systems based on accurate maps of their infrastructure. Asset detection is an important tool towards automation of processes and improved decision support on such systems. Due to lack of available data, limited research exists investigating railway asset detection, despite the rise of Artificial Neural Networks and the numerous investigations on autonomous driving. Here, we present a novel dataset used in real world projects for mapping railway assets. Also, we implement Faster R-CNN, a state of the art deep learning object detection method, for detection of signs and signals on this dataset. We achieved 79.36% on detection and a 70.9% mAP. The results were compromised by the small size of the objects, the low resolution of the images and the high similarity across classes.


Railway Object detection Object recognition Deep learning Faster R-CNN 


  1. 1.
    Agudo, D., Sánchez, Á., Vélez, J.F., Moreno, A.B.: Real-time railway speed limit sign recognition from video sequences. In: 2016 International Conference on Systems, Signals and Image Processing (IWSSIP), pp. 1–4, May 2016Google Scholar
  2. 2.
    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, vol. 2015, pp. 91–99 (2015)Google Scholar
  3. 3.
    Zheng, D., Wang, Y.: Application of an artificial neural network on railway passenger flow prediction. In: Proceedings of 2011 International Conference on Electronic Mechanical Engineering and Information Technology, vol. 1, pp. 149–152, August 2011Google Scholar
  4. 4.
    Tsai, T.-H., Lee, C.-K., Wei, C.-H.: Neural network based temporal feature models for short-term railway passenger demand forecasting. Expert Syst. Appl. 36(2), 3728–3736 (2009). Part 2. Scholar
  5. 5.
    Karthick, N., Nagarajan, R., Suresh, S., Prabhu, R.: Implementation of railway track crack detection and protection. Int. J. Eng. Comput. Sci. 6(5), 21476–21481 (2017). Scholar
  6. 6.
    Sadeghi, J., Askarinejad, H.: Application of neural networks in evaluation of railway track quality condition. J. Mech. Sci. Technol. 26(1), 113–122 (2012). Scholar
  7. 7.
    Gibert, X., Patel, V.M., Chellappa, R.: Robust fastener detection for autonomous visual railway track inspection. In: 2015 IEEE Winter Conference on Applications of Computer Vision, pp. 694–701, January 2015Google Scholar
  8. 8.
    Sinha, D., Feroz, F.: Obstacle detection on railway tracks using vibration sensors and signal filtering using bayesian analysis. IEEE Sens. J. 16(3), 642–649 (2016)CrossRefGoogle Scholar
  9. 9.
    Faghih-Roohi, S., Hajizadeh, S., Núñez, A., Babuska, R., Schutter, B.D.: Deep convolutional neural networks for detection of rail surface defects. In: 2016 International Joint Conference on Neural Networks (IJCNN), pp. 2584–2589, July 2016Google Scholar
  10. 10.
    Marmo, R., Lombardi, L., Gagliardi, N.: Railway sign detection and classification. In: 2006 IEEE Intelligent Transportation Systems Conference, pp. 1358–1363, September 2006Google Scholar
  11. 11.
    Melander, M., Halme, I.: Computer vision based solution for sign detection. Eur. Railway Rev. (2016).
  12. 12.
    Arastounia, M.: Automated recognition of railroad infrastructure in rural areas from LIDAR data. Remote Sens. 11, 14916–14938 (2015)CrossRefGoogle Scholar
  13. 13.
    Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 580–587 (2014)Google Scholar
  14. 14.
    Girshick, R.: Fast R-CNN. In: Proceedings (IEEE International Conference on Computer Vision), vol. 2015, pp. 1440–1448 (2015)Google Scholar
  15. 15.
    He, K., Gkioxari, G., Dollár, P., Girshick, R.: Mask R-CNN. In: Proceedings of the International Conference on Computer Vision (ICCV) (2017)Google Scholar
  16. 16.
    Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: unified, real-time object detection. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. 2016, pp. 779–788 (2016)Google Scholar
  17. 17.
    Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.-Y., Berg, A.C.: SSD: Single shot multibox detector. In: Lecture Notes in Computer Science, vol. 9905, pp. 21–37 (2016)Google Scholar
  18. 18.
    Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollar, P., Zitnick, C.L.: Microsoft COCO: common objects in context. In: Lecture Notes in Computer Science, vol. 8693, pp. 740–755 (2014)Google Scholar
  19. 19.
    Great Britain: Department for Transport. Traffic Signs Manual. The Stationary Office, London, United Kingdom (2013)Google Scholar
  20. 20.
    Safety, R., Board, S.: Lineside Operational Safety Signs. Railway Group Standard, London (2009)Google Scholar
  21. 21.
    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. 2016, pp. 770–778 (2016)Google Scholar
  22. 22.
    Everingham, M., Gool, L., Williams, C.K., Winn, J., Zisserman, A.: The pascal visual object classes (VOC) challenge. Int. J. Comput. Vis. 88(2), 303–338 (2010). Scholar
  23. 23.
    Salton, G., McGill, M.J.: Introduction to Modern Information Retrieval. McGraw-Hill Inc., New York (1986)zbMATHGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Georgios Karagiannis
    • 1
    • 2
    Email author
  • Søren Olsen
    • 1
  • Kim Pedersen
    • 1
  1. 1.Department of Computer ScienceUniversity of CopenhagenCopenhagenDenmark
  2. 2.COWI A/SLyngbyDenmark

Personalised recommendations