Character Prototyping in Document Images Using Gabor Filters

  • Bénédicte Allier
  • Hubert Emptoz
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2749)


In this article we present a particular application of Gabor filtering for machine-printed document image understanding. To do so, we assume that the text can be seen as texture, characters being the smallest texture elements, and we verify this hypothesis by a series of experiments over different sets of character images. We first apply a bank of 24 Gabor filters (4 frequencies and 6 orientations) on each set, then we extract texture features, that are used to classify character images without a priori knowledge using a Bayesian classifier. Results are shown for different characters written in a same font, and for different font types given a character.


  1. 1.
    Chaudhuri, B. B. et Garain, U. Extraction of type style-based meta-information from imaged documents. IJDAR. Vol. 3(3) (2001) 138–149.CrossRefGoogle Scholar
  2. 2.
    Wong, K. Y., Casey, R. G. et Wahl, F. M. Document Analysis System. IBM Journal of Research and Development. Vol. 26(6) (1982) 647–656.CrossRefGoogle Scholar
  3. 3.
    Eglin, V., Bres, S. et Emptoz, H. Statistical characterization and classification of printed text in a multiscale context. Proc. of 2nd Int. Workshop on SPR, Sydney (NSW), (Australia), (1998) 960–967.Google Scholar
  4. 4.
    Duffy, L. Recherche d’information logique dans les documents à typographie riche et récurrente. Application aux sommaires. Thesis of the INSA de Lyon, Lyon (France) (1997).Google Scholar
  5. 5.
    Doermann, D. S., Rivlin, E. et Rosenfeld, A. The function of documents. IJCV. Vol. 16(11) (1998) 799–814.Google Scholar
  6. 6.
    Jain, A. K., Bhattacharjee, S. K. et Chen, Y. On texture in document images. Proc. of IEEE CVPR, Champaign, Illinois (USA), (1992) 677–680.Google Scholar
  7. 7.
    Wu, V., Manmatha, R. et Riseman, E. M. Finding Text in Images. Proc. of ACM Int. Conf. on Digital Libraries, Philadelphia, PA (USA), (1997) 23–26.Google Scholar
  8. 8.
    Zhu, Y., Tan, T. et Wang, Y. Font Recognition Based On Global Texture Analysis. IEEE Trans. on PAMI. Vol. 23(10) (2001) 1192–1200.Google Scholar
  9. 9.
    Porat, M. et Zeevi, Y. Y. The generalized Gabor scheme of image representation in biological and machine vision. IEEE Trans. on PAMI. Vol. 10(4) (1988) 452–468.zbMATHGoogle Scholar
  10. 10.
    Manjunath, B. S. et Ma, W. Y. Texture Features for Browsing and Retrieval of Image Data. IEEE Trans. on PAMI. Vol. 18(8) (1996) 837–842.Google Scholar
  11. 11.
    Cheeseman, P. et Stutz, J. Bayesian classification (Autoclass): Theory and results. Adv. In Knowledge Discovery And Data Mining (1996) 153–180.Google Scholar
  12. 12.
    Allier, B. et Boursier, N. Reconstruction de caractères dégradés: étude pour la détermination d’un caractère “idéal”. Lyon (FRANCE): Institut National des Sciences Appliquées-INSA; Report No.: RR2002-05 (décembre 2002).Google Scholar
  13. 13.
    Chenevoy, Y. Reconnaissance structurelle de documents imprimés: études et réalisations. Thesis of the INPL, (1992).Google Scholar
  14. 14.
    Zramdini, A. Study of Optical Font Recognition Based on Global Typographical Features. Thesis of the Université de Fribourg (Suisse) (1995).Google Scholar
  15. 15.
    Jain, A. K. et Farrokhnia, F. Unsupervised Texture Segmentation Using Gabor Filters. Pattern Recognition. Vol. 24(12) (1991) 1167–1186.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Bénédicte Allier
    • 1
  • Hubert Emptoz
    • 1
  1. 1.Reconnaissance des Formes et Vision Laboratory (RFV)INSA LyonVilleurbanne cedexFrance

Personalised recommendations