People Detection and Pose Classification Inside a Moving Train Using Computer Vision

  • Sergio A. VelastinEmail author
  • Diego A. Gómez-Lira
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10645)


The use of surveillance video cameras in public transport is increasingly regarded as a solution to control vandalism and emergency situations. The widespread use of cameras brings in the problem of managing high volumes of data, resulting in pressure on people and resources. We illustrate a possible step to automate the monitoring task in the context of a moving train (where popular background removal algorithms will struggle with rapidly changing illumination). We looked at the detection of people in three possible postures: Sat down (on a train seat), Standing and Sitting (half way between sat down and standing). We then use the popular Histogram of Oriented Gradients (HOG) descriptor to train Support Vector Machines to detect people in any of the predefined postures. As a case study, we use the public BOSS dataset. We show different ways of training and combining the classifiers obtaining a sensitivity performance improvement of about 12% when using a combination of three SVM classifiers instead of a global (all classes) classifier, at the expense of an increase of 6% in false positive rate. We believe this is the first set of public results on people detection using the BOSS dataset so that future researchers can use our results as a baseline to improve upon.


People detection Posture classification People monitoring On-board surveillance Machine learning 



The work described here was carried out as part of the OBSERVE project funded by the Fondecyt Regular Program of Conicyt (Chilean Research Council for Science and Technology) under grant no. 1140209. S.A. Velastin is grateful to funding received from the Universidad Carlos III de Madrid, the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no. 600371, el Ministerio de Economía y Competitividad (COFUND2013-51509) and Banco Santander.


  1. 1.
    La Tercera: Cámaras de seguridad en la Región Metropolitana aumentarán en un 78% (2010). Accessed 24 June 2017
  2. 2.
    Evans, I.: Report: London no safer for all its CCTV cameras (2012). Accessed 24 June 2017
  3. 3.
    BC News: 1,000 cameras ‘solve one crime’ (2009). Accessed 24 June 2017
  4. 4.
    Burges, C.J.: A tutorial on support vector machines for pattern recognition. Data Min. knowl. Discov. 2(2), 121–167 (1998)CrossRefGoogle Scholar
  5. 5.
    Nowozin, S., Lampert, C.H.: Structured learning and prediction in computer vision. Found. Trends® Comput. Graph. Vis. 6(3), 185–365 (2011)zbMATHGoogle Scholar
  6. 6.
    Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition CVPR 2005, vol. 1, pp. 886–893 (2005)Google Scholar
  7. 7.
    Chen, G., Hou, R.: A new machine double-layer learning method and its. In: International Conference on Mechatronics and Automation ICMA 2007, pp. 796–799 (2007)Google Scholar
  8. 8.
    Wang, Z., Yoon, S., Hong, C., Park, D.S.: A novel SVM based pedestrian detection algorithm via locality sensitive histograms. In: Proceedings of the International Conference on Image Processing, Computer Vision, and Pattern Recognition (IPCV), vol. 2, p. 1, The Steering Committee of The World Congress in Computer Science, Computer Engineering and Applied Computing (WorldComp) (2014)Google Scholar
  9. 9.
    Wang, Z., Yoon, S., Xie, S.J., Lu, Y., Park, D.S.: A high accuracy pedestrian detection system combining a cascade AdaBoost detector and random vector functional-link net. Sci. World J. 2014, 7 p. (2014). doi: 10.1155/2014/105089. Article ID 105089
  10. 10.
    Cuyang, W., Wang, X.: Joint deep learning for pedestrian detection. In: 2013 IEEE International Conference on Computer Vision (ICCV), pp. 2056–2063. IEEE (2013)Google Scholar
  11. 11.
    Zeng, X., Ouyang, W., Wang, M., Wang, X.: Deep learning of scene-specific classifier for pedestrian detection. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014 Part III. LNCS, vol. 8691, pp. 472–487. Springer, Cham (2014). doi: 10.1007/978-3-319-10578-9_31 Google Scholar
  12. 12.
    Fukui, H., Yamashita, T., Yamauchi, Y., Fujiyoshi, H., Murase, H.: Pedestrian detection based on deep convolutional neural network with ensemble inference network. In: 2015 IEEE Intelligent Vehicles Symposium (IV), pp. 223–228. IEEE (2018)Google Scholar
  13. 13.
  14. 14.
    Cong, D.N.T., Achard, C., Khoudour, L.: People re-identification by classification of silhouettes based on sparse representation. In: 2010 2nd International Conference on Image Processing Theory Tools and Applications (IPTA), pp. 60–65. IEEE (2010)Google Scholar
  15. 15.
    Cong, D.N.T., Khoudour, L., Achard, C., Flancquart, A.: Adaptive model for object detection in noisy and fast-varying environment. In: Image Analysis and Processing–ICIAP 2011, pp. 68–77 (2011)Google Scholar
  16. 16.
    Coniglio, C., Meurie, C., Lézoray, O., Berbineau, M.: People silhouette extraction from people detection bounding boxes in images. Pattern Recog. Lett. 93, 182–191 (2017)CrossRefGoogle Scholar
  17. 17.
    University of Maryland.: ViPER: The Video Performance Evaluation Resource (2003). Accessed 24 June 2017
  18. 18.
    Quinteros, D., Velastin, S.A., Acuna, G.: Characterisation of the spatial sensitivity of classifiers in pedestrian detection. In: 6th LatinAmerican Conference on Networked Electronic Media, Medellin, Colombia (2015)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Department of Computer ScienceUniversidad Carlos III de MadridColmenarejoSpain
  2. 2.Queen Mary University of LondonLondonUK
  3. 3.Department of Informatics EngineeringUniversidad de Santiago de ChileSantiagoChile

Personalised recommendations