Modeling, Evaluation and Control of a Road Image Processing Chain

  • Yves Lucas
  • Antonio Domingues
  • Driss Driouchi
  • Pierre Marché
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3540)


Tuning a complete image processing chain (IPC) remains a tricky step. Until now researchers focused on the evaluation of single algorithms, based on a small number of test images and ad hoc tuning independent of input data. In this paper we explain how, by combining statistical modeling with design of experiments, numerical optimization and neural learning, it is possible to elaborate a powerful and adaptive IPC. To succeed, it is necessary to build a large image database, to describe input images and finally to evaluate the IPC output. By testing this approach on an IPC dedicated to road obstacle detection, we demonstrate that this experimental methodology and software architecture ensure a steady efficiency. The reason is simple: the IPC is globally optimized, from a large number of real images (180 out of a sequence of 30 000) and with adaptive processing of input data.


Input Image Optimal Tuning Homogeneity Measure Large Image Database Adaptive Tuning 
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.


  1. 1.
    Haralick, R.M.: Performance characterization protocol in computer vision. In: ARPA Image Understanding Workshop, Monterey, CA, pp. 667–673 (1994)Google Scholar
  2. 2.
    Courtney, P., Thacker, N., Clark, A.: Algorithmic modeling for performance evaluation. In: Workshop on perf. characteristics of vision algorithms, Cambridge, April 19, p. 13 (1996)Google Scholar
  3. 3.
    Forstner, W.: 10 Pros and cons against performance characterization of vision algorithms. In: Workshop on Performance Characteristics of Vision Algorithms (April 1996)Google Scholar
  4. 4.
    Bowyer, K.W., Phillips, P.J.: Empirical Evaluation Techniques in Computer Vision, p. 262. Wiley-IEEE Computer Society Press (June 1998) (ISBN: 0-8186-8401-1)Google Scholar
  5. 5.
    Meer, P., Matei, B., Cho, K.: Input Guided Performance Evaluation. Theoretical Foundations of Computer Vision, 115–124 (1998)Google Scholar
  6. 6.
    Phillips, I.T., Chhabra, A.K.: Empirical Performance Evaluation of Graphics Recognition Systems. IEEE PAMI 21, 849–870 (1999)Google Scholar
  7. 7.
    Blanc-Talon, J., Ropert, V.: Evaluation des chaînes de traitement d’images. Revue Scientifique et Technique de la Défense, vol. 46, pp. 29–38 (2000)Google Scholar
  8. 8.
    Philipp-Foliguet, S.: Evaluation de la segmentation, ETIS, Cergy-Pontoise, Mars (2001)Google Scholar
  9. 9.
    Sebe, N., Tian, Q., Loupias, E., Lew, M., Huang, T.: Evaluation of salient point techniques. In: Lew, M., Sebe, N., Eakins, J.P. (eds.) CIVR 2002. LNCS, vol. 2383, p. 367. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  10. 10.
    Rosin, P., Ioannidis, E.: Evaluation of global image thresholding for change detection. Pattern Recognition Letters 24, 2345–2356 (2003)zbMATHCrossRefGoogle Scholar
  11. 11.
    Yitzhaky, Y., Peli, E.: A method for objective edge detection evaluation and detection parameter selection. IEEE PAMI 25(8), 1027–1033 (2003)Google Scholar
  12. 12.
    Ropert, V.: Proposition d’une architecture de contrôle pour un système de vision, Thèse de l’Université René Descartes (Paris 6), France, Décembre (2001)Google Scholar
  13. 13.
    Levner, I., Bulitko, V.: Machine learning for adaptive image Interpretation. In: The Sixteenth Innovative Appl. of Art. Intell. Conf (IAAI 2004), San Jose (CA) USA, July 27-29, p. 7 (2004)Google Scholar
  14. 14.
    Schimmerling, P., Sisson, J.-C., Zaïdi, A.: Pratique des Plans d’Expériences., Lavoisier Tec & Doc (1998), ISBN 2-743-00239-5Google Scholar
  15. 15.
    Treuillet, S.: Analyse de l’influence des paramètres d’une chaîne de traitements d’images par un plan d’expériences. In: 19e colloque GRETSI 2003, September 8-11 (2003)Google Scholar
  16. 16.
    Margaret, H.: Wright The Nelder-Mead Simplex Method. In: Recent Theory and Practice Int. Symposium on Mathematical Programming Lausanne, EPFL, August 24-29 (1997)Google Scholar
  17. 17.
    Domingues, A., Lucas, Y., Baudrier, D., Marché, P.: Détection et suivi d’objets en temps Détection et suivi d’objets en temps réel par un système embarqué multi capteurs. In: GRETSI 2001, Toulouse (September 2001)Google Scholar
  18. 18.
    Domingues, A.: Système embarqué multicapteurs pour la détection d’obstacles routiers - Développement du prototype et réglage automatique de la chaîne de traitement d’images, Thèse de l’Université d’Orléans, France 15 Juillet (2004)Google Scholar
  19. 19.
    Lucas, Y., Domingues, A., Boubal, M., Marché, P.: Système de vision embarqué pour la détection d’obstacles routiers Techniques de l’Ingénieur – Recherche Innovation (2005)Google Scholar
  20. 20.
    Lamaty, P.: Opérateurs de niveau intermédiaire pour le traitement temps réel des images, Thèse de Doctorat, France (2000)Google Scholar
  21. 21.
    Fries, A., Hunter, J.: Minimum aberration 2k − p designs. Technometrics 22, 601–608 (1980)zbMATHCrossRefMathSciNetGoogle Scholar
  22. 22.
    Rechtschaffner, R.L.: Saturated fractions of 2n and 3n factorial designs. Technometrics 9, 569–575 (1967)CrossRefMathSciNetGoogle Scholar
  23. 23.
    Cheng, H., Sun, Y.: A hierarchical approach to color image segmentation using homogeneity. IEEE transactions on image processing 9(12), 2071–2082 (2000)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Yves Lucas
    • 1
  • Antonio Domingues
    • 2
  • Driss Driouchi
    • 2
  • Pierre Marché
    • 3
  1. 1.Vision and Robotics Lab, IUT Mesures PhysiquesOrleans UniversityBourges cedexFrance
  2. 2.Vision and Robotics LabENSI of BourgesBourgesFrance
  3. 3.Theoretical and Applied Statistics LabPierre & Marie Curie UniversityParisFrance

Personalised recommendations