Advertisement

Evolution of Computer Vision Subsystems in Robot Navigation and Image Classification Tasks

  • Sascha Lange
  • Martin Riedmiller
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3276)

Abstract

Real-time decision making based on visual sensory information is a demanding task for mobile robots. Learning on high-dimensional, highly redundant image data imposes a real problem for most learning algorithms, especially those being based on neural networks. In this paper we investigate the utilization of evolutionary techniques in combination with supervised learning of feedforward nets to automatically construct and improve suitable, task-dependent preprocessing layers helping to reduce the complexity of the original learning problem. Given a number of basic, parameterized low-level computer vision algorithms, the proposed evolutionary algorithm automatically selects and appropriately sets up the parameters of exactly those operators best suited for the imposed supervised learning problem.

Keywords

Good Individual Training Pattern Locally Linear Embedding Linear Embedding Control Subsystem 
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.

References

  1. 1.
    Bala, J., DeJong, K., Huang, J., Vafaie, H., Wechsler, H.: Hybrid Learning Using Genetic Algorithms and Decision Tress for Pattern Classification. In: 14th Int. Joint Conf. on Artifical Intelligence (IJCAI), Canada, pp. 719–724 (1995)Google Scholar
  2. 2.
    Belpaeme, T.: Evolution of Visual Feature Detectors. In: Proceedings of the First European Workshop on Evolutionary Computation in Image analysis and Signal Processing (EvoIASP 1999, Göteborg, Sweden), University of Birmingham School of Computer Science technical report (1999)Google Scholar
  3. 3.
    Braun, H., Weisbrod, J.: Evolving feedforward neural networks. In: Proc. of the International Conference on Artificial Neural Nets and Genetic Algorithms (1993)Google Scholar
  4. 4.
    Braun, H., Ragg, T.: Evolutionary Optimization of Neural Networks for Reinforcement Learning Algorithms. In: ICML 1996, Workshop Proceedings on Evolutionary Computing and Machine Learning, Italy, pp. 38–45 (1996)Google Scholar
  5. 5.
    McCabe, G.P.: Principal Variables. Technometrics 26, 127–134 (1984)CrossRefMathSciNetGoogle Scholar
  6. 6.
    Draper., B.: Learning Object Recognition Strategies. Ph.D. dissertation, Univ. of Massachusetts, Dept. of Computer Science. Tech. report 93–50 (1993)Google Scholar
  7. 7.
    Draper, B.: Learning Control Strategies for Object Recognition. In: Ikeuchi, Veloso (eds.) Visual Learning. Oxford University Press, Oxford (1996)Google Scholar
  8. 8.
    Jolliffe, I.T.: Principal Component Analysis. Springer, Heidelberg (1986)Google Scholar
  9. 9.
    Krzanowski, W.J.: Selection of Variables to Preserve Multivariate Data Structure, Using Principal Component Analysis. Applied Statistics- Journal of the Royal Statistical Society Series C 36, 22–33 (1987)Google Scholar
  10. 10.
    Lange, S.: Verfolgung von farblich markierten Objekten in 2 Dimensionen. B.Sc. thesis, Universität Osnabrück, Institut für Kognitionswissenschaft (2001)Google Scholar
  11. 11.
    Martin, C.M.: The Simulated Evolution of Robot Perception. Ph.D. dissertation, Carnegie Mellon University Pittsburgh (2001)Google Scholar
  12. 12.
    Priese, L., Rehrmann, V., Schian, R., Lakmann, R.: Traffic Sign Recognition Based on Color Image Evaluation. In: Proc. Intelligent Vehicles Symposium, pp. 95–100 (1993)Google Scholar
  13. 13.
    De Ridder, D., Kouropteva, O., Okun, O., Pietikäinen, M., Duin, R.P.W.: Supervised locally linear embedding. In: Kaynak, O., Alpaydın, E., Oja, E., Xu, L. (eds.) ICANN 2003 and ICONIP 2003. LNCS, vol. 2714, pp. 333–341. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  14. 14.
    Riedmiller, M., Braun, H.: A direct adaptive method for faster backpropagation learning: the Rprop algorithm. In: Proceedings of the ICNN, San Francisco (1993)Google Scholar
  15. 15.
    Roweis, S.T., Saul, L.K.: Nonlinear dimensionality reduction by locally linear embedding. Science 290(5500), 2323–2326 (2000)CrossRefGoogle Scholar
  16. 16.
    Smith, S.M.: A new class of corner finder. In: Proc. 3rd British Machine Vision Conference, pp. 139–148 (1992)Google Scholar
  17. 17.
    Steels, L., Kaplan, F.: AIBO’s first words. The social learning of language and meaning. In: Gouzoules, H. (ed.) Evolution of Communication, vol. 4(1). John Benjamins Publishing Company, Amsterdam (2001)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Sascha Lange
    • 1
  • Martin Riedmiller
    • 1
  1. 1.Neuroinformatics Group, Institute for Computer Science and Institute for Cognitive ScienceUniversity of OsnabrückOsnabrückGermany

Personalised recommendations