Line and cell searching in tables or forms

  • E. Turolla
  • Y. Belaïd
  • A. Belaïd
Pattern Recognition and Document Processing
Part of the Lecture Notes in Computer Science book series (LNCS, volume 974)


Item searching is an important step in form analysis. The goal of this paper is to describe a robust method to locate the items whose boundaries are black lines, and an algorithm to classify those items. Our method is based on detection of lines by Hough transform and on searching of cycles in a graph that represents the cells (boundaries of the items). We will show that our approach is robust, skew independent and can be applied to several kind of lines such as continuous, dashed, doubled, etc. We classify the items in four categories: blank, black, gray and meaningful.


Hough Transform French Tax Forms Line Searching Item Extraction Item Classification 


  1. [1]
    D. S. DOERMANN, A. ROSENFELD, “the Processing of Form Documents”, ICDAR 1993, pp. 497–501Google Scholar
  2. [2]
    J. YUAN, L. XU, C.Y. SUEN, “Form Items Extraction by Model matching”, ICDAR 1991, pp.210–218.Google Scholar
  3. [3]
    G. MADERLECHNER, “'Symbolic subtraction’ of fixed formatted graphics and text from filled in forms”, MVA'90, IAPR Workshop on Machine Vision Applications, Nov 28–30 1990, TokyoGoogle Scholar
  4. [4]
    R. KASTURI, R. RAMAN, C. CHENNUBHOTLA, L. O'GORMAN “Document image analysis, an overview of techniques for graphics recognition”, SSPR 1990.Google Scholar
  5. [5]
    T. RISSE, “Hough Transform for Line Recognition: Complexity of Evidence Accumulation and Cluster Detection”, Computer Vision, Graphics, and Image Processing, vol. 46, pp. 327–345, 1989Google Scholar
  6. [6]
    Y. MULLER-BELAÏD, R. MOHR “Planes and quadrics detection using Hough transform”, 7th International Conference on Pattern Recognition; Montreal, Canada, July 30–August 2, 1984Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • E. Turolla
    • 1
  • Y. Belaïd
    • 1
  • A. Belaïd
    • 1
  1. 1.CRIN-CNRS, Bât. LoriaVandoeuvre-lès-Nancy CedexFrance

Personalised recommendations