Object-Level Fusion and Confidence Management in a Multi-Sensor Pedestrian Tracking System

Chapter
First Online:
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 35)

Abstract

This paper 1 describes a multi-sensor fusion system dedicated to detect, recognize and track pedestrians. The fusion by tracking method is used to fuse asynchronous data provided by different sensors with complementary and supplementary fields of view. Having the performance of the sensors, we propose to differentiate between the two problems: object detection and pedestrian recognition, and to quantify the confidence in the detection and recognition processes. This confidence is calculated based in geometric features and it is updated under the Transferable Belief Model framework. The vehicle proprioceptive data are filtered by a separate Kalman filter and are used in the estimation of the relative and absolute state of detected pedestrians. Results are shown with simulated data and with real experimental data acquired in urban environment.

Keywords

Keywords Multi-sensor data fusion Pedestrians’ detection and recognition Transferable belief Model Confidence management 
This is a preview of subscription content, log in to check access.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Sensors and system architecture for vulnerable road users protection, http://www.save-u.org/
  2. 2.
    Integrated solutions for preventive and active safety systems, http://www.prevent-ip.org/
  3. 3.
    Cooperative vehicles and road infrastructure for road safety, http://www.safespot-eu.org/
  4. 4.
    Cherfaoui V., Denoeux T., and Cherfi Z-L. Distributed data fusion: application to the confidence management in vehicular networks. Proceedings of the 11th International Conference on Information Fusion, Cologne, Germany, pp. 846–853 , July 2008.Google Scholar
  5. 5.
    Fuerstenberg K.C. and Lages U. Pedestrian detection and classification by Laser scanners. 9th EAEC International Congress, Paris, France, June 2003.Google Scholar
  6. 6.
    Suard F. et al. Pedestrian detection using stereo-vision and graph kernels. Proceedings of the IEEE Intelligent Vehicles Symposium IV’2005, pp. 267–272, June 2005.Google Scholar
  7. 7.
    Viola P., Jones M.J., and Snow D. Detecting pedestrians using patterns of motion and appearance. Proceedings of 9th IEEE International Conference on the Computer Vision, Vol. 63, pp. 153–161, Nice, France, October 2003.Google Scholar
  8. 8.
    Tessier C. et al. A real-time, multi-sensor architecture for fusion of delayed observations: application to vehicle localization. IEEE ITSC’2006, Toronto, Canada, September 2006.Google Scholar
  9. 9.
    Sittler R.W. An optimal data association problem in surveillance theory. IEEE Transactions on Military Electronics, 8(2), 125–139, 1964.CrossRefGoogle Scholar
  10. 10.
    Fayad F. and Cherfaoui V. Tracking objects using a laser scanner in driving situation based on modeling target shape, Proceedings of the IEEE Intelligent Vehicles Symposium IV’2007, pp. 44–49 Istanbul, Turkey, June 2007.Google Scholar
  11. 11.
    Bonnifait Ph. et al. Dynamic localization of car-like vehicle using data fusion of redundant ABS sensors. The Journal of Navigation, 56, 429–441, 2003.CrossRefGoogle Scholar
  12. 12.
    Julier S. and Durrant-Whyte H. Process models for the high-speed navigation of road vehicles. Proceedings of the IEEE International Conference on Robotics and Automation, Vol. 1, pp. 101-105, Nagoya, Japan, May 1995.Google Scholar
  13. 13.
    Fayad F., Cherfaoui V., and Dherbomez G. Updating confidence indicators in a multi-sensor pedestrian tracking system. Proceedings of the IEEE Intelligent Vehicles Symposium IV’2008, Eindhoven, Holland, June 2008.Google Scholar
  14. 14.
    Smets Ph. and Kennes R. The transferable belief model. Artificial Intelligence, 66(2), 191–234, 1994.MATHCrossRefMathSciNetGoogle Scholar
  15. 15.
    Denoeux T. The cautious rule of combination for belief functions and some extensions. Proceedings of the 9th International Conference on Information Fusion, Florence, Italy, pp. 1–8, July 2006.Google Scholar
  16. 16.
    Denoeux T. Conjunctive and disjunctive combination of belief functions induced by non distinct bodies of evidence. Artificial Intelligence, 172, 234–264, 2008.CrossRefMathSciNetGoogle Scholar
  17. 17.
    Elouedi Z. et al. Assessing sensor reliability for multisensor data fusion within the transferable belief model. IEEE Transactions on Systems, Man and Cybernetics, 34, 782, 2004.CrossRefGoogle Scholar
  18. 18.
    Sudano J.J. Inverse pignistic probability transforms. Proceedings of the 5th International Conference on Information Fusion, Vol. 2, pp. 763–768, 2002.Google Scholar
  19. 19.
    Fayad F. and Cherfaoui V. Detection and recognition confidences update in a multi-sensor pedestrian tracking system. Proceedings of the 12th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems IPMU2008, Malaga, Spain, June 2008.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  1. 1.Heudiasyc UMR CNRS 6599, Université de Technologie de Compiégne, Centre de Recherche de Royallieu – BP 20529Compiégne CedexFrance

Personalised recommendations