Advertisement

Float Cascade Method for Pedestrian Detection

  • Yanwen Chong
  • Qingquan Li
  • Hulin Kuang
  • Chun-Hou Zheng
Part of the Communications in Computer and Information Science book series (CCIS, volume 304)

Abstract

A novel pedestrian detection method based on the Four Direction Features (FDF), called FloatCascade pedestrian detection, is proposed for the pedestrian detection problem, which can be applied to the pedestrian detection problem in a single image. The FDF can represent pedestrian well, and the computation cost is lower than the HOG’s. FloatCascade applies the plus-l-minus-r method to select the effective cascade features to improve the detection performance, where l is fixed by experience, r is a float value which must be lower than l or quite to l. whether add features or not is decided by the detection rate, while whether subtract features or not is decided by error rate. Experimental results show that the MutualCascade method is more effective than Voila and Jones’ cascade and some other Adaboost-based method, and is comparable with HOG-based methods. It also demonstrates a higher performance compared with the state-of-the-art methods.

Keywords

FDF GAB CAdaboost FloatCascade pedestrian detection 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Dollar, P., Wojek, C., Schiele, B., Perona, P.: Pedestrian Detection: Evaluation of the State of the Art. IEEE Transactions on Pattern Analysis and Machine Intelligence 30(2), 60–80 (2011)Google Scholar
  2. 2.
    Xu, R., Jiao, J., Zhang, B., Ye, Q.: Pedestrian Detection in Images via Cascaded L1 Norm Minimization Learning Method. Pattern Recognition 45(3), 2573–2583 (2012)CrossRefGoogle Scholar
  3. 3.
    Soga, M., Hiratsuka, S.: Pedestrian Detection for a Near Infrared Imaging System. In: Proc. the 11th International IEEE Conference on Intelligent Transportation Systems, pp. 12–15 (2008)Google Scholar
  4. 4.
    Viola, P.A., Jones, M.J.: Robust Real-Time Face Detection. Intl. Journal of Computer Vision 57(2), 137–154 (2004)CrossRefGoogle Scholar
  5. 5.
    Li, S., Zhang, Z.: FloatBoost Learning and Statistical Face Detection. IEEE Trans. Pattern Analysis and Machine Intelligence 26(9), 1112–1123 (2004)CrossRefGoogle Scholar
  6. 6.
    Kittler, J.: Feature Set Search Algorithm. Pattern Recognition in Practice, 41–60 (1980)Google Scholar
  7. 7.
    Pudil, P., Novovicova, J., Choakjarernwanit, N., Kittler, J.: A Comparative Evaluation of Floating Search Methods for Feature Selection. Technical Report VSSP-TR-5/92, University of Surrey, UK (1992)Google Scholar
  8. 8.
    Dalal, N., Triggs, B.: Histograms of Oriented Gradients for Human Detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, vol. 1(3), pp. 886–893 (2005)Google Scholar
  9. 9.
    Zhu, Q., Avidan, S., Yeh, M., Cheng, K.T.: Fast Human Detection Using a Cascade of Histograms of Oriented Gradients. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, vol. 2(3), pp. 1491–1498 (2006)Google Scholar
  10. 10.
    Tuzel, Q., Porikli, F., Meer, P.: Human Detection via Classification on Riemannian Manifolds. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–8 (2007)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Yanwen Chong
    • 1
    • 2
  • Qingquan Li
    • 1
    • 2
  • Hulin Kuang
    • 1
    • 2
    • 3
  • Chun-Hou Zheng
    • 4
  1. 1.State Key Laboratory for Information Engineering in Surveying, Mapping and Remote SensingWuhan UniversityWuhanChina
  2. 2.Engineering Research Center for Spatio-Temporal Data Smart Acquisition and ApplicationMinistry of Education of ChinaWuhanChina
  3. 3.School of Electronic InformationWuHan UniversityWuhanChina
  4. 4.College of Electrical Engineering and AutomationAnhui UniversityHefeiChina

Personalised recommendations