A Hybrid Approach to Detect Graphical Symbols in Documents

  • Salvatore Tabbone
  • Laurent Wendling
  • Daniel Zuwala
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3163)


We propose to combine a feature descriptor method with a structural representation of symbols. An adaptation of the Radon transform, keeping main geometric transformations usually required for the recognition of symbols, is provided. In order to improve the recognition step we directly process on the grey level document. In this perspective, a three-dimensional signature integrates into a same formalism both the shape of the object and its photometric variations. More precisely the signature is computed within the symbol following several grey levels. Additionally a structural representation of symbols allows to localize into the document candidate symbols.


Structural Representation Junction Point Symbol Recognition Graphic Recognition Skeleton Image 
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.
    Cordella, L.P., Vento, M.: Symbol recognition in documents: a collection of techniques? International Journal on Document Analysis and Recognition 3, 73–88 (2000)CrossRefGoogle Scholar
  2. 2.
    Belkasim, S.O., Shridar, M., Ahmadi, M.: Pattern Recognition with Moment Invariants: A Comparative Study and New Results. Pattern Recognition 24, 1117–1138 (1991)CrossRefGoogle Scholar
  3. 3.
    Belongie, S., Puzicha, J.M., Shape, J.: matching and object recognition using shape contexts. IEEE Transactions on PAMI 24, 509–522 (2002)Google Scholar
  4. 4.
    Ghorbel, F.: A complete invariant description for gray level images by harmonic analysis approach. Pattern Recognition Letters 15, 1043–1051 (1994)CrossRefGoogle Scholar
  5. 5.
    Lin, B.C., Shen, J.: Fast Computation of Moment Invariants. Pattern Recognition 24, 807–813 (1991)CrossRefGoogle Scholar
  6. 6.
    Llado, J., Jose, E.M., Symbol Recognition, J.: by Subgraph Matching Between Region Adjancy Graphs. IEEE Transactions on PAMI 23, 1137–1143 (2001)Google Scholar
  7. 7.
    Messmer, B., H.B.: Automatic learning and recognition of graphical symbols in engineering drawings. In: Kasturi, R., Tombre, K. (eds.) Graphics Recognition 1995. LNCS, vol. 1072, pp. 123–134. Springer, Heidelberg (1996)Google Scholar
  8. 8.
    B.G. Park, K.M. Lee, S.L., Lee, J.: Recognition of partially occluded objects using probabilistic ARG (attributed relational graph)-based matching. Computer Vision and Image Understanding 90, 217–241 (2003)Google Scholar
  9. 9.
    Ramel, J.-Y., Boissier, G., Emptoz, H.: A structural representation adapted to handwritten symbol recognition. In: Chhabra, A.K., Dori, D. (eds.) GREC 1999. LNCS, vol. 1941, pp. 259–266. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  10. 10.
    Yan, L., Wenyin, L.: Engineering drawings recognition using a case-based approach. International Conference on Document Analysis and Recognition, Edinburgh 2886, 190–194 (2003)Google Scholar
  11. 11.
    Arcelli, C., di Baja, G.S.: A One-Pass Two-Operation Process to Detect the Skeletal Pixels on the 4-Distance Transform. IEEE Transactions on PAMI 11, 411–414 (1989)Google Scholar
  12. 12.
    Deans, S.R.: Applications of the Radon Transform. Wiley Interscience Publications, New York (1983)zbMATHGoogle Scholar
  13. 13.
    Tabbone, S., Wendling, L.: Binary shape normalization using the radon transform. In: Nyström, I., Sanniti di Baja, G., Svensson, S. (eds.) DGCI 2003. LNCS, vol. 2886, pp. 184–193. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  14. 14.
    Borgefors, G.: An Improved Version of the Chamfer Matching Algorithm. In: Proceedings of 7th International Conference on Pattern Recognition, Montréal (Canada), vol. 2, pp. 1175–1177 (1984)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Salvatore Tabbone
    • 1
  • Laurent Wendling
    • 1
  • Daniel Zuwala
    • 1
  1. 1.Loria-UMR 7503, Campus ScientifiqueVillers-lès-Nancy CedexFrance

Personalised recommendations