Skip to main content

Advertisement

SpringerLink
Log in

Constructing a pedestrian recognition system with a public open database, without the necessity of re-training: an experimental study

  • Theoretical Advances
  • Published:
Pattern Analysis and Applications Aims and scope Submit manuscript

Abstract

Pedestrian recognition is one of the basic elements of an active pedestrian protection system. Especially, there have been various researches to develop vision-based pedestrian classifiers. Recently, Munder and Gavrila (IEEE Trans Pattern Anal Mach Intell 28(11):1863–1868, 2006) opened their DaimlerChrysler (DCX) pedestrian image database and tried to provide an objective comparison between popular features and classifiers. After their publication, objective performance comparison of features, classifiers and architectures became possible. This paper reports four experimental results with the DCX database. First, the Gabor filter bank-based feature is competent in pedestrian recognition. Second, optimization of a classifier using performance estimator greatly enhances the performance of the resultant classifier. Third, once the imaging system uses histogram equalization and the same capturing method as a database, a practical pedestrian recognition system can be constructed with a public open database, without the necessity of re-training using the new database acquired with the actual imaging system. Fourth, a posteriori probability-based post-processing increases the recognition rate of consecutive image sequencing while maintaining a false positive rate.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

References

  1. Gandhi T, Trivedi MM (2007) Pedestrian protection systems: issues, survey, and challenges. IEEE Trans Pattern Anal Mach Intell 8(3):413–430

    Google Scholar 

  2. Munder S, Gavrila DM (2006) An experimental study on pedestrian classification. IEEE Trans Pattern Anal Mach Intell 28(11):1863–1868

    Article  Google Scholar 

  3. Liang Y-h, Wang Z-y, Guo S, Xu X-w, Cao X-y (2007) Pedestrian detection using KPCA and FLD algorithms. In: Proceedings of the IEEE international conference on automation and logistics. Jinan, China, 18–21 August 2007, pp 1572–1575

  4. Jia H-X, Zhang Y-J (2007) Fast human detection by boosting histogram of oriented gradients. In: 9th biennial conference of the Australian pattern recognition society on digital image computing techniques and applications. 3–5 December 2007, pp 638–688

  5. Paisitkriangkrai S, Shen C, Zhang J (2007) An experimental evaluation of local features for pedestrian classification. In: 9th biennial conference of the Australian pattern recognition society on digital image computing techniques and applications. 3–5 December 2007, pp 53–60

  6. Strackeljan J, Ghio A (2009) NiSIS Competition 2007. NiSIS (Nature-inspired Smart Information Systems). http://www.nisis.risk-technologies.com/filedown.aspx?file=274

  7. NiSIS (2009) NiSIS 2007 program and papers. In: NiSIS 2007—3rd annual symposium. St. Julians, Malta, 26–27 November 2007

  8. Movellan JR (2006) Tutorial on Gabor filters. http://mplab.ucsd.edu/tutorials/pdfs/gabor.pdf

  9. Manjunath BS, Ma WY (1998) Texture features for browsing and retrieval of image data. IEEE Trans Pattern Anal Mach Intell 18(8):837–842

    Article  Google Scholar 

  10. Liu C, Wechsler H (2003) Independent component analysis of Gabor features for face recognition. IEEE Trans Neural Netw 14(4):919–928

    Article  Google Scholar 

  11. Sun Z, Bebis G, Miller R (2006) Monocular precrash vehicle detection: features and classifiers. IEEE Trans Image Process 15(7):2019–2034

    Article  Google Scholar 

  12. Jung HG, Lee YH, Yoon PJ, Hwang IY, Kim J (2006) Sensor fusion-based obstacle detection/classification for active pedestrian protection system. Lect Notes Comput Sci 4292:294–305

    Article  Google Scholar 

  13. Burges CJC (1998) A tutorial on support vector machines for pattern recognition. Data Min Knowl Discov 2:121–167

    Article  Google Scholar 

  14. Chen D, Cao XB, Xu YW, Qiao H, Wang FY (2006) SVM-based classifier with shape and motion features for a pedestrian detection system. In: Intelligent vehicle symposium 2006. Tokyo, Japan, 13–15 June 2006, pp 331–335

  15. Yuan X, Sun Z, Varol Y, Bebis G (2003) A distributed visual surveillance system. In: Proceedings of the IEEE conference on advanced video and signal based surveillance (AVSS’03), 21–22 July 2003, pp 199–204

  16. Jung HG, Yoon PJ, Kim J (2007) Genetic algorithm-based optimization of SVM-based pedestrian classifier. In: The 22nd international technical conference on circuits/systems, computers and communications. July 2007, pp 783–784

  17. Schauland S, Kummert A, Park S-B, Iurgel U, Zhang Y (2006) Vision-based pedestrian detection—improvement and verification of feature extraction methods and SVM-based classification. In: Proceedings of the IEEE intelligent transportation system conference. Toronto, Canada, 17–20 September 2006

  18. Joachims T (2000) Estimating the generalization performance of an SVM efficiently. In: Proceedings of the international conference on machine learning. Mogan Kaufman Publishers Inc., 2000, pp 431–438

  19. Chen D, Cao XB, Xu YW, Qiao H (2006) An evolutionary support vector machines classifier for pedestrian detection. In: Proceedings of the 2006 IEEE/RSJ international conference on intelligent robots and systems. Beijing, China, 9–15 October 2006, pp 4223–4227

  20. Thorsten Joachims (2009) SVMlight. http://svmlight.joachims.org/

  21. Munder S (2007) Multi-cue pedestrian recognition. PhD dissertation, Mannheim University, 2007

  22. Center for Biological and Computational Learning (CBCL) at MIT (2009) CBCL Pedestrian Database #1. http://cbcl.mit.edu/software-datasets/PedestrianData.html

  23. MANDO Corporation (2009) MANDO pedestrian image database. http://web.yonsei.ac.kr/hgjung/APPS

  24. Gavrila DM, Letellier L, Meinecke M-M (2003) SAVE-U Deliverable 1-B: vulnerable road user sensor data collection. http://www.save-u.org

  25. Gavrila DM, Marchal PL, Meinecke M-M (2003) SAVE_U Deliverable 1-A: vulnerable road user scenario analysis. http://www.save-u.org

  26. Gavrila DM (2003) A Bayesian, exemplar-based approach to hierarchical shape matching. IEEE Trans Pattern Anal Mach Intell 29(8):1408–1421

    Article  Google Scholar 

  27. Ogawa G, Kise K, Torii T, Nagao T (2007) Onboard evolutionary risk recognition system for automobiles—toward the risk map system. IEEE Trans Ind Electron 54(2):878–886

    Article  Google Scholar 

  28. Cao X-B, Qiao H, Keane J (2008) A low-cost pedestrian-detection system with a single optical camera. IEEE Trans Intell Transp Syst 9(1):1–10

    Article  Google Scholar 

  29. De Nicolao G, Ferrara A, Giacomini L (2007) Onboard sensor-based collision risk assessment to improve pedestrian safety. IEEE Trans Veh Technol 56(5):2405–2413

    Article  Google Scholar 

  30. Cheng H, Zheng N, Zhang X, Qin J, van de Wetering H (2007) Interactive Road Situation Analysis for Driver Assistance and Safety Warning Systems: Framework and Algorithms. IEEE Transactions on Intelligent Transportation Systems 8(1):157–167

    Article  Google Scholar 

  31. Broggi A, Cerri P, Ghidoni S, Grisleri P, Jung HG (2008) Localization and analysis of critical areas in urban scenarios. In: IEEE intelligent vehicle symposium. Eindhoven University of Technology, Eindhoven, The Netherlands, 4–6 June 2008, pp 1074–1079

Download references

Acknowledgments

We wish to thank the authors of [2, 21], Munder and Gavrila, for providing ROC data and dissertation. And, we also wish to thank the authors of [5], Paisitkriangkrai et al., for providing detailed experimental results.

Author information

Authors and Affiliations

  1. MANDO Corporation, Global R&D H.Q., 413-5, Gomae-Dong, Giheung-Gu, Yongin-Si, Kyonggi-Do, 446-901, Korea

    Ho Gi Jung

  2. School of Electrical and Electronic Engineering, Yonsei University, 134 Sinchon-Dong, Seodaemun-Gu, Seoul, 120-749, Korea

    Jaihie Kim

Authors
  1. Ho Gi Jung
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jaihie Kim
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Ho Gi Jung.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Jung, H.G., Kim, J. Constructing a pedestrian recognition system with a public open database, without the necessity of re-training: an experimental study. Pattern Anal Applic 13, 223–233 (2010). https://doi.org/10.1007/s10044-009-0153-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10044-009-0153-2

Keywords

Search

Navigation