Learning for feature selection and shape detection

  • Rita Cucchiara
  • Massimo Piccardi
  • Michele Bariani
  • Paola Mello
Poster Session B: Active Vision, Motion, Shape, Stereo
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1310)


The paper proposes a general framework for shape detection based on supervised symbolic learning. Differently from other visual systems exploiting machine learning, the proposed architecture does not follow the object segmentation - feature extraction and (learning based) classification approach. Instead, an initial data-driven processing selects points of interest in the scene by means of complex features which hypothesize the presence of the target shape; hypotheses are validated by a classifier defined by a machine learning algorithm. Learning is exploited not only for defining the model, i.e. the description of the target for the classifier, but also for defining the description language, i.e. the feature set useful in generating reliable object hypotheses. The proposed architecture of visual system has been implemented for an industrial application of unstructured shape detection: examples and results are reported in the paper


  1. 1.
    Cho, K., Dunn, S. M.: Learning shape classes. IEEE Trans. on PAMI 16 (1994) n. 9 882–887Google Scholar
  2. 2.
    Drapter, B. A., Brodley, C. E., Utgoff, P. E.: Goal directed classification using Linear Machine Decision tree. IEEE Trans. on PAMI 16 (1994) n. 9 888–893Google Scholar
  3. 3.
    Pellegretti, P., Roli, F., Serpico, S., Vernazza G.: Supervised learning of descriptions for image recognition purposes. IEEE Trans. on PAMI 16 (1994) n. 1 92–98Google Scholar
  4. 4.
    Murase, H., Nayar, S. K.: Learning by a generation approach to appearance based object recognition. Proc. of 13th ICPR, Vienna 1 (1996) 24–30Google Scholar
  5. 5.
    Cucchiara, R., Filicori, F., Andreetta, R.: Detecting micro-cracks in ferromagnetic material with automatic visual inspection. Proc. of QCAV95 Int. Conf. on Quality Control by Artificial Vision, Le Creusot France, (1995) 19–24Google Scholar
  6. 6.
    Bolle, R. M., Califano, A., Kjeldsen, R.: A complete and extendable approach to visual recognition. IEEE Trans. on PAMI 14 (1992) n. 5 534–548Google Scholar
  7. 7.
    Newman, T. S., Jain, A.K.: A Survey of automated visual inspection. Comp. Vision and Image understanding 61 (1995) n. 2 231–262Google Scholar
  8. 8.
    Illingworth, J., Kittler, J.: A Survey of the Hough transform. Comput. Vision Graphics, Image Process. 43 (1988) 221–238Google Scholar
  9. 9.
    Cucchiara, R., Piccardi, M.: Eliciting visual primitives for detecting elongated shapes, 3rd Int. Workshop on Visual Form Capri, Italy, (1997) (to appear)Google Scholar
  10. 10.
    Quinlan, J. R.: C4.5 Programs for machine learning. Morgan Kaufmann Publ. (1993)Google Scholar
  11. 11.
    Bariani, M., Cucchiara, R., Piccardi, M., Mello, P.: Data mining for automated visual inspection, Proc. of PADD97, London, UK, (1997) 51–64Google Scholar
  12. 12.
    Haralick, R., Shapiro, L.: Computer and Robot Vision vol. I, Addison Wesley (1991)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Rita Cucchiara
    • 1
  • Massimo Piccardi
    • 1
  • Michele Bariani
    • 1
  • Paola Mello
    • 1
  1. 1.Dipartimento di Ingegneria University of FerraraFerraraItaly

Personalised recommendations