Advertisement

Gait Anti-spoofing

  • John D. BustardEmail author
  • Mohammad Ghahramani
  • John N. Carter
  • Abdenour Hadid
  • Mark S. Nixon
Chapter
Part of the Advances in Computer Vision and Pattern Recognition book series (ACVPR)

Abstract

Gait recognition is a relatively new biometric and as a result relatively little effort has yet been devoted to studying spoofing attacks against it. This chapter examines the effects of two different spoofing attacks against two different gait recognition systems. The first attack uses clothing impersonation where an attacker replicates the clothing of a legitimately enrolled individual. The second attack is a targeted attack where an imposter deliberately selects the legitimately enrolled subject whose gait signature is closest to the attacker. The analysis presented here reveals that both systems are vulnerable to both attacks. In particular, if both attacks are combined and the systems have acceptance thresholds set at the EER of their baseline performance, the attacks cause the FAR to rise from  5 % to between 60 and 95 %. The chapter describes two countermeasures that can be applied to minimise the effects of the spoofing attacks. Using the same acceptance thresholds the countermeasure to clothing attacks reduces the FAR performance under clothing impersonation from 40 to 15 %. Likewise, the targeting countermeasure reduces the FAR for targeted attacks from 20 to 2.5 % sufficient to even improve on the baseline performance.

Keywords

Gait Cycle Equal Error Rate Replay Attack Baseline System False Acceptance Rate 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Notes

Acknowledgments

The authors would like to thank the Academy of Finland and the TABULA RASA project (http://www.tabularasa-euproject.org) funded under the 7th Framework Programme of the European Union (EU) (grant agreement number 25728) for their financial support.

References

  1. 1.
    Nixon M, Carter J (2006) Automatic recognition by Gait. Proc IEEE 94(11):2013–2024CrossRefGoogle Scholar
  2. 2.
    Nixon MS, Tan TN, Chellappa R (2005) Human identification based on Gait. International series on biometrics. Springer, New YorkGoogle Scholar
  3. 3.
    Hadid A, Ghahramani M, Kellokumpu V, Pietikainen M, Bustard J, Nixon M (2012) Can gait biometrics be spoofed? In: 21st International Conference on Pattern Recognition (ICPR ), pp 3280–3283 2012Google Scholar
  4. 4.
    Kellokumpu V, Zhao G, Li SZ, Pietikainen M (2009) Dynamic texture based gait recognition. 3rd IAPR/IEEE International conference on biometrics, pp 1000–1009 2009Google Scholar
  5. 5.
    Kellokumpu V, Zhao G, Pietikäinen M (2010) Dynamic textures for human movement recognition. ACM International conference on image and video retrieval, CIVR ’10, 470–476 2010Google Scholar
  6. 6.
    Seely RD, Samangooei S, Middleton L, Carter J, Nixon M (2008) The University of Southampton multi-biometric tunnel and introducing a novel 3d gait dataset. In: Biometrics: theory, applications and systems. IEEE, 2008Google Scholar
  7. 7.
    Sarkar S, Phillips PJ, Liu Z, Vega IR, Grother P, Bowyer KW (2005) The humanid gait challenge problem: data sets, performance, and analysis. IEEE Trans Pattern Anal Mach Intell 27:162–177CrossRefGoogle Scholar
  8. 8.
    Seely RD (2010) On a three-dimensional gait recognition system. Ph.D. thesis, University of SouthamptonGoogle Scholar
  9. 9.
    Matovski D, Nixon M, Mahmoodi S, Carter J (2011) The effect of time on gait recognition performance. IEEE Trans Inf Forensics Secur 7(2):543–552CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London 2014

Authors and Affiliations

  • John D. Bustard
    • 1
    Email author
  • Mohammad Ghahramani
    • 2
  • John N. Carter
    • 1
  • Abdenour Hadid
    • 2
  • Mark S. Nixon
    • 1
  1. 1.University of SouthamptonSouthamptonUK
  2. 2.Pentti Kaiteran katu 1OuluFinland

Personalised recommendations