Multimedia Tools and Applications

, Volume 75, Issue 5, pp 2473–2486 | Cite as

Visual railway detection by superpixel based intracellular decisions



Safety is one of the most concerned issues in traffic and transportation, among which railway detection is a fundamental and necessary research. In this paper, we propose a visual railway detection method based on superpixels rather than pixels. An SVM classifier is learned based on features, on which a TF-IDF like transform is applied, and it greatly improves the performance of the classification. The intracellular decision scheme is proposed to make decisions on a superpixel by using predictions of features within the superpixel. All the superpixels that are predicted as positive constitute the railway to be detected. The proposed railway detection method is evaluated on a number of challenging images and experiments demonstrate that the proposed method is an effective and detailed solution to railway detection, and is superior to other railway detection methods.


Railway detection Superpixel Intracellular decision scheme 



This work was supported by the Fundamental Research Funds for the Central Universities with grant number 2014JBM040.


  1. 1.
    Achanta R, Shaji A, Smith K, Lucchi A, Fua P, Susstrunk S (2010) SLIC Superpixels. EPFL Technical Report 149300Google Scholar
  2. 2.
    Alexe B, Deselaers T, Ferrari V (2010) What is an object? IEEE Conf. on Computer Vision and Pattern Recognition (CVPR). p 73–80Google Scholar
  3. 3.
    Chang C-C, Lin C-J (2011) LIBSVM: a library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2:27:1--27:27. Software available at
  4. 4.
    Dalal N, Triggs B (2005) Histograms of oriented gradients for human detection. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR). 886–893Google Scholar
  5. 5.
    De Ruvo P, De Ruvo G, Distante A, Nitti M, Stella E, Marino F (2008) A visual inspection system for rail detection & tracking in real time railway maintenance. The Open Cybernetics and Systemics Journal 2:57–67CrossRefGoogle Scholar
  6. 6.
    Esveld C (2001) Modern railway track, (Second Edition), MRT-ProductionsGoogle Scholar
  7. 7.
    Felzenszwalb PF, Girshick RB, McAllester D, Ramanan D (2010) Object detection with discriminatively trained part-based models. IEEE Trans Pattern Anal Machine Intell 32(9):1627–1645CrossRefGoogle Scholar
  8. 8.
    Fulkerson B, Vedaldi A, Soatto S (2008) Localizing objects with smart dictionaries. Proceedings of the European Conference on Computer Vision. p 179–192Google Scholar
  9. 9.
    Garcia J, Losada C, Espinosa F, Ureña J, Hernández Á, Mazo M, Jiménez A, Bueno E, De Marziani C, Álvarez F (2005) Dedicated smart IR barrier for obstacle detection in railways. IEEE Industrial Electronics Society, IECONGoogle Scholar
  10. 10.
    Jaakkola TS, Haussler D (1998) Exploiting generative models in discriminative classifiers. In Advances in Neural Information Processing Systems 11Google Scholar
  11. 11.
    Jun W, Shen H, Li Y-D, Xiao Z-B, Ming-Yu L, Wang C-L (2013) Learning a hybrid similarity measure for image retrieval. Pattern Recogn 46(11):2927–2939CrossRefMATHGoogle Scholar
  12. 12.
    Kaleli F, Akgul YS (2009) Vision-based railroad track extraction using dynamic programming. Proceedings of the 12th Int. IEEE Conference on Intelligent Transportation SystemsGoogle Scholar
  13. 13.
    Kang D-J, Jung M-H (2003) Road lane segmentation using dynamic programming for active safety vehicles. Pattern Recogn Lett 24:3177–3185CrossRefGoogle Scholar
  14. 14.
    Kato T, Ninomiya Y, Masaki I (2002) An obstacle detection method by fusion of radar and motion stereo. IEEE Trans Intell Transp Syst 3:182–188CrossRefGoogle Scholar
  15. 15.
    Kobayashi T (2014) Dirichlet-based histogram feature transform for image classification. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR). 4321–4328Google Scholar
  16. 16.
    Lang C, Liu G, Yu J, Yan S (2012) Saliency detection by multitask sparsity pursuit. IEEE Trans Image Process 21(3):1327–1338MathSciNetCrossRefGoogle Scholar
  17. 17.
    Li Y, Shen H (2013) On identity disclosure control for hypergraph-based data publishing. IEEE TransInf Forensics Security 8(8):1384–1396CrossRefGoogle Scholar
  18. 18.
    Lowe DG (1999) Object recognition from local scale-invariant features. Proceedings of the International Conference on Computer Vision 2. 1150–1157Google Scholar
  19. 19.
    Maire F, Bigdeli A (2010) Obstacle-free range determination for rail track maintenance vehicles. 11th International conference on Control Automation Robotics Vision (ICARCV). 2172–2178Google Scholar
  20. 20.
    Mockel, F. Scherer, P F. Schuster (2003) Multi-sensor obstacle detection on railway tracks. Proc. IEEE Int. Intelligent Vehicles Symp p 42–46Google Scholar
  21. 21.
    Nassu BT, Ukai M (2011) Rail extraction for driver support in railways. IEEE Intelligent Vehicles Symposium (IV) Baden-Baden, GermanyGoogle Scholar
  22. 22.
    Papageorgiou CP, Oren M, Poggio T (1998) A general framework for object detection. Proceedings of the International Conference on Computer Vision. p 555–562Google Scholar
  23. 23.
    Passarella R, Tutuko B, Prasetyo APP (2011) Design concept of train obstacle detection system in Indonesia. IJRRAS 9(3)Google Scholar
  24. 24.
    Qi Z, Tian Y, Shi Y (2013) Efficient railway tracks detection and turnouts recognition method using HOG features. Neural Comput & Applic 23:245–254CrossRefGoogle Scholar
  25. 25.
    Reisert M, Burkhardt H (2008) Equivariant holomorphic filters for contour denoising and rapid object detection. IEEE Trans Image Processing 17(2):190–203MathSciNetCrossRefGoogle Scholar
  26. 26.
    Robertson S (2004) Understanding inverse document frequency: on theoretical arguments for IDF. J Doc 60(5):503–520CrossRefGoogle Scholar
  27. 27.
    Ruder M, Mohler N, Ahmed F (2003) An obstacle detection system for automated trains. Proceedings of the Intelligent Vehicles Symposium. 180–185Google Scholar
  28. 28.
    Singh M, Singh S, Jaiswal J, Hempshall J (2006) Autonomous rail track inspection using vision based system. IEEE International Conference on Computational Intelligence for Homeland Security and Personal Safety AlexandriaGoogle Scholar
  29. 29.
    Song Z, Chen Q, Huang Z, Hua Y, Yan S (2011) Contextualizing object detection and classification. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR). 1585–1592Google Scholar
  30. 30.
    Sparck Jones K (1972) A statistical interpretation of term specificity and its application in retrieval. J Doc 28:11–21CrossRefGoogle Scholar
  31. 31.
    Vedaldi A, Fulkerson B (2008) Vlfeat: an open and portable library of computer vision algorithms. ACM International Conference on Multimedia.
  32. 32.
    Wang T, Dai G, Ni B, De X, Siewe F (2012) A distance measure between labeled combinatorial maps. Comput Vis Image Underst 116:1168–1177CrossRefGoogle Scholar
  33. 33.
    Wohlfeil J (2011) Vision based rail track and switch recognition for self-localization of trains in a rail network. IEEE Intelligent Vehicles Symposium (IV), Baden-Baden, GermanyGoogle Scholar
  34. 34.
    Zhang Z, Weiss R, Hanson AR (1994) Qualitative obstacle detection. Tech Report COMPSI TR94-20Google Scholar

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  1. 1.School of Computer and Information TechnologyBeijing Jiaotong UniversityBeijingChina

Personalised recommendations