Logo LVSN
EnglishAccueil
A proposPersonnesRecherchePublicationsEvenementsProfil
A propos
Publications

 

 

 

 

REPARTI

MIVIM

A Fuzzy-Syntactic Approach to Allograph Modeling for Cursive Script Recognition


Marc Parizeau and Réjean Plamondon


Abstract - On-line cursive script recognition has recently become the focus of renewed interest because of the development of notebook computers that incorporate a digitizing tablet over a high-resolution graphics display. This new hardware, incorporating the electronic penpad concept, aims at the elimination of the keyboard and mouse for many interactive applications. However, although recognition algorithms for isolated characters are widely used for this purpose, cursive script recognition is just starting to be proposed in commercial systems and still remains a stumbling block. The object of this paper is to present an original method for creating allograph models and recognizing them within cursive handwriting. This method concentrates on the morphological aspect of cursive script recognition. It uses fuzzy-shape grammars to define the morphological characteristics of conventional allographs which can be viewed as basic (a priori) knowledge for developing a multi-writer recognition system. The system developed uses no linguistic knowledge to output character sequences that possibly correspond to an unknown cursive word input. The recognition method is tested using multi-writer cursive random letter sequences. For a test dataset containing a handwritten cursive text 600 characters in length written by ten different writers, average character recognition rates of 84.4% to 91.6% are obtained, depending on whether only the first (best) character sequence output of the system is considered or if the best of the top ten is accepted. These results are achieved without any writer-dependent tuning. The same dataset is used to evaluate the performance of human readers. An average recognition rate of 96.0% was reached, using ten different readers, presented with randomized samples of each writer. The worst reader-writer performance was 78.3%. Moreover, results show that system performances are highly correlated with human performances.

download document

Bibtex:

@article{Parizeau61,
    author    = { Marc Parizeau and Réjean Plamondon },
    title     = { A Fuzzy-Syntactic Approach to Allograph Modeling for Cursive Script Recognition },
    volume    = { 17 },
    number    = { 7 },
    pages     = { 702-712 },
    year      = { 1995 },
    month     = { July },
    journal   = { IEEE Trans. on Pattern Analysis and Machine Intelligence }
}

Dernière modification: Jun 14 2002 5:04PM par parizeau

     
   
   

©2002-. Laboratoire de Vision et Systèmes Numériques. Tous droits réservés