Semantic Segmentation Based Traffic Light Detection at Day and at Night

  • Vladimir Haltakov
  • Jakob Mayr
  • Christian Unger
  • Slobodan Ilic
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9358)


Traffic light detection from a moving vehicle is an important technology both for new safety driver assistance functions as well as for autonomous driving in the city. In this paper we present a machine learning framework for detection of traffic lights that can handle in real-time both day and night situations in a unified manner. A semantic segmentation method is employed to generate traffic light candidates, which are then confirmed and classified by a geometric and color features based classifier. Temporal consistency is enforced by using a tracking by detection method.

We evaluate our method on a publicly available dataset recorded at daytime in order to compare to existing methods and we show similar performance. We also present an evaluation on two additional datasets containing more than 50 intersections with multiple traffic lights recorded both at day and during nighttime and we show that our method performs consistently in those situations.


  1. 1.
    Traffic Safety Facts 2008. National Highway Traffic Safety Administration (2008)Google Scholar
  2. 2.
    Fachserie. 8, Verkehr. 7, Verkehrsunfälle. Statistisches Bundesamt Wiesbaden (2013)Google Scholar
  3. 3.
    Cai, Z., Li, Y., Gu, M.: Real-time recognition system of traffic light in urban environment. In: CISDA (2012)Google Scholar
  4. 4.
    de Charette, R., Nashashibi, F.: Real time visual traffic lights recognition based on spot light detection and adaptive traffic lights templates. In: IV (2009)Google Scholar
  5. 5.
    Chiang, C.C., Ho, M.C., Liao, H.S., Pratama, A., Syu, W.C.: Detecting and recognizing traffic lights by genetic approximate ellipse detection and spatial texture layouts. Int. J. Innovative Comput. Inf. Control 17, 6919–6934 (2011)Google Scholar
  6. 6.
    Diaz-Cabrera, M., Cerri, P.: Traffic light recognition during the night based on fuzzy logic clustering. In: Moreno-Díaz, R., Pichler, F., Quesada-Arencibia, A. (eds.) EUROCAST. LNCS, vol. 8112, pp. 93–100. Springer, Heidelberg (2013) CrossRefGoogle Scholar
  7. 7.
    Diaz-Cabrera, M., Cerri, P., Sanchez-Medina, J.: Suspended traffic lights detection and distance estimation using color features. In: ITS (2012)Google Scholar
  8. 8.
    Fairfield, N., Urmson, C.: Traffic light mapping and detection. In: ICRA (2011)Google Scholar
  9. 9.
    Fan, B., Lin, W., Yang, X.: An efficient framework for recognizing traffic lights in night traffic images. In: CISP (2012)Google Scholar
  10. 10.
    Franke, U., Pfeiffer, D., Rabe, C., Knoeppel, C., Enzweiler, M., Stein, F., Herrtwich, R.G.: Making bertha see. In: ICCV Workshop on Computer Vision for Autonomous Driving (2013)Google Scholar
  11. 11.
    Gomez, A., Alencar, F., Prado, P., Osorio, F., Wolf, D.: Traffic lights detection and state estimation using hidden markov models. In: IV (2014)Google Scholar
  12. 12.
    Gong, J., Jiang, Y., Xiong, G., Guan, C., Tao, G., Chen, H.: The recognition and tracking of traffic lights based on color segmentation and camshift for intelligent vehicles. In: IV (2010)Google Scholar
  13. 13.
    Haltakov, V., Unger, C., Ilic, S.: Geodesic pixel neighborhoods for multi-class image segmentation. In: BMVC (2014)Google Scholar
  14. 14.
    Hel-Or, Y., Hel-Or, H.: Real time pattern matching using projection kernels. In: ICCV (2003)Google Scholar
  15. 15.
    Insurance Institute for Highway Safety (IIHS): Status Report, vol. 42, no. 1. Rep. IIHS (2007).
  16. 16.
    Jang, C., Kim, C., Kim, D., Lee, M., Sunwoo, M.: Multiple exposure images based traffic light recognition. In: IV (2014)Google Scholar
  17. 17.
    Kim, H.K., Shin, Y.N., Kuk, S.G., Park, J.H., Jung, H.Y.: Night-time traffic light detection based on SVM with geometric moment features. World Acad. Sci. Eng. Technol. 7, 454–457 (2013)Google Scholar
  18. 18.
    Kim, Y., Kim, K., Yang, X.: Real time traffic light recognition system for color vision deficiencies. In: ICMA (2007)Google Scholar
  19. 19.
    Li, J.: An efficient night traffic light recognition method. J. Inf. Comput. Sci. 10(9), 2773–2781 (2013)CrossRefGoogle Scholar
  20. 20.
    Lindner, F., Kressel, U., Kaelberer, S.: Robust recognition of traffic signals. In: IV (2004)Google Scholar
  21. 21.
    Omachi, M., Omachi, S.: Traffic light detection with color and edge information. In: ICCSIT (2009)Google Scholar
  22. 22.
    Shen, Y., Ozguner, U., Redmill, K., Liu, J.: A robust video based traffic light detection algorithm for intelligent vehicles. In: IV (2009)Google Scholar
  23. 23.
    Siogkas, G., Skodras, E., Dermatas, E.: Traffic lights detection in adverse conditions using color, symmetry and spatiotemporal information. In: VISAPP 2012 (2012)Google Scholar
  24. 24.
    Tae-Hyun, H., In-Hak, J., Seong-Ik, C.: Detection of traffic lights for vision-based car navigation system. In: Chang, L.-W., Lie, W.-N. (eds.) PSIVT 2006. LNCS, vol. 4319, pp. 682–691. Springer, Heidelberg (2006) CrossRefGoogle Scholar
  25. 25.
    Torralba, A., Murphy, K.P., Freeman, W.T.: Sharing visual features for multiclass and multiview object detection. PAMI 29(5), 854–869 (2007)CrossRefGoogle Scholar
  26. 26.
    Wang, C., Jin, T., Yang, M., Wang, B.: Robust and real-time traffic lights recognition in complex urban environments. Int. J. Comput. Intell. Syst. 4(6), 1383–1390 (2011)CrossRefGoogle Scholar
  27. 27.
    Wojek, C., Schiele, B.: A dynamic conditional random field model for joint labeling of object and scene classes. In: Forsyth, D., Torr, P., Zisserman, A. (eds.) ECCV 2008, Part IV. LNCS, vol. 5305, pp. 733–747. Springer, Heidelberg (2008) CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Open Access This chapter is distributed under the terms of the Creative Commons Attribution Noncommercial License, which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.

Authors and Affiliations

  • Vladimir Haltakov
    • 1
    • 2
  • Jakob Mayr
    • 1
    • 3
  • Christian Unger
    • 1
    • 2
  • Slobodan Ilic
    • 2
  1. 1.BMW GroupMunichGermany
  2. 2.Technical University MunichMunichGermany
  3. 3.Munich University of Applied SciencesMunichGermany

Personalised recommendations