Elsevier

Pattern Recognition

Volume 30, Issue 3, March 1997, Pages 421-445
Pattern Recognition

Word shape analysis for a hybrid recognition system

https://doi.org/10.1016/S0031-3203(96)00093-3Get rights and content

Abstract

This paper describes two wholistic recognizers developed for use in a hybrid recognition system. The recognizers use information about the word shape. This information is strongly related to word zoning. One of the recognizers is explicitly limited by the accuracy of the zoning information extraction. The other recognizer is designed so as to avoid this limitation. The recognizers use very simple sets of features and fuzzy set based pattern matching techniques. This not only aims to increase their robustness, but also causes problems with disambiguation of the results. A verification mechanism, using letter alternatives as compound features, is introduced. Letter alternatives are obtained from a segmentation based recognizer coexisting in the hybrid system. Despite some remaining disambiguation problems, wholistic recognizers are found capable of outperforming the segmentation based recognizer. When working together in a hybrid system, the results are significantly higher than that of the individual recognizers. Recognition results are reported and compared.

References (24)

  • W. Guerfali et al.

    Normalizing and restoring on-line handwriting

    Pattern Recognition

    (1993)
  • J.S. Lipscomb

    A trianable gesture recognizer

    Pattern Recognition

    (1991)
  • R.K. Powalka et al.

    Multiple word segmentation with interactive look-up for cursive script recognition

  • R.K. Powalka et al.

    Dynamic cursive script recognition. A hybrid approach

  • T.K. Ho et al.

    A regression approach to combination of decisions by multiple character recognition algorithms

  • B. Plessis et al.

    A multi-classifier combination strategy for the recognition of handwritten cursive words

  • L. Xu et al.

    Methods for combining multiple classifiers and their applications to handwriting recognition

    IEEE Trans. Systems Man Cybernet.

    (1992)
  • L.D. Earnest

    Machine recognition of cursive writing, Information Processing

  • R.F.H. Farag

    Word-level recognition of cursive script

    IEEE Trans. Comput.

    (1979)
  • J.-C. Simon

    Off-line cursive word

  • M.K. Brown et al.

    Cursive script recognition

  • L.A. Zadeh

    Fuzzy sets

    Inf. Control

    (1965)
  • Cited by (20)

    • A pictorial dictionary for printed Farsi subwords

      2008, Pattern Recognition Letters
    • Holistic cursive word recognition based on perceptual features

      2007, Pattern Recognition Letters
      Citation Excerpt :

      Guillevic et al. (1995) takes into account the number of strokes crossing within the main body of the word. Powalka et al. (1997) estimate the word length based on the number of times the stroke crosses the horizontal axis. However, word length cannot be estimated precisely enough before word recognition, even when using robust estimation methods.

    • Lexicon reduction of handwritten Arabic subwords based on the prominent shape regions

      2016, International Journal on Document Analysis and Recognition
    • Handwritten word recognition using Web resources and recurrent neural networks

      2015, International Journal on Document Analysis and Recognition
    • Lexicon reduction for printed Farsi subwords using pictorial and textual dictionaries

      2014, International Journal on Document Analysis and Recognition
    View all citing articles on Scopus
    View full text