Automated Reading of Cheque Amounts

Abstract: This paper describes a complete system for the automated reading of amounts extracted from bank and postal cheques. We focus our description on the steps where we developed and applied new techniques. These are the reading of the legal amount and the cross-validation of the recognition results for the legal and courtesy amount. In our work we deal with German cheque amounts. The reading of German legal amounts is a special challenge, since the literal words the legal amount is constructed of are usually completely connected. For the cross-validation step, we present a well structured combination strategy that allows detecting and correcting recognition errors for the legal and courtesy amount and leads to a significant increase in the system performance. Our system was tested on a database with real cheques from Swiss postal services.

[1]  Horst Bunke,et al.  Error Localization and Correction in Check Processing , 1999 .

[2]  Heinrich Niemann,et al.  Automatic speech recognition without phonemes , 1993, EUROSPEECH.

[3]  H. Niemann,et al.  ISADORA | a Speech Modelling Network Based on Hidden Markov Models , 1993 .

[4]  Christopher Raphael,et al.  Language-Independent and Segmentation-Free Techniques for Optical Character Recognition , 1996, DAS.

[5]  Torsten Caesar,et al.  Preprocessing and feature extraction for a handwriting recognition system , 1993, Proceedings of 2nd International Conference on Document Analysis and Recognition (ICDAR '93).

[6]  Ching Y. Suen,et al.  Bank check processing system , 1996 .

[7]  Alexander Filatov,et al.  Check Amount Recognition Based on the Cross Validation of Courtesy and Legal Amount Fields , 1997 .

[8]  Horst Bunke,et al.  Amount translation and error localization in check processing using syntax-directed translation , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[9]  Horst Bunke,et al.  Detection and correction of recognition errors in check reading , 2000, International Journal on Document Analysis and Recognition.

[10]  Yves Lecourtier,et al.  Multi-Bank Check Recognition System: Consideration on The Numeral Amount Recognition Module , 1997, Int. J. Pattern Recognit. Artif. Intell..

[11]  Horst Bunke,et al.  Automatic bankcheck processing , 1997 .

[12]  Horst Bunke,et al.  Off-line handwritten numeral string recognition by combining segmentation-based and segmentation-free methods , 1998, Pattern Recognit..

[13]  Edouard Lethelier,et al.  An automatic reading system for handwritten numeral amounts on French checks , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[14]  L. Rabiner,et al.  An introduction to hidden Markov models , 1986, IEEE ASSP Magazine.

[15]  Edouard Lethelier,et al.  Automatic Reading of Handwritten Amounts on French Checks , 1997, Int. J. Pattern Recognit. Artif. Intell..

[16]  Sebastiano Impedovo,et al.  Automatic Bankcheck Processing: A New Engineered System , 1997, Int. J. Pattern Recognit. Artif. Intell..

[17]  Rolf-Dieter Bippus 1-dimensional and pseudo 2-dimensional HMMs for the recognition of German literal amounts , 1997, Proceedings of the Fourth International Conference on Document Analysis and Recognition.

[18]  Gyeonghwan Kim,et al.  An architecture for handwritten text recognition systems , 1999, International Journal on Document Analysis and Recognition.

[19]  Mohamed Cheriet,et al.  Finding ascenders and descenders of handwritten words , 1990 .

[20]  Horst Bunke,et al.  Off-line cursive handwriting recognition using hidden markov models , 1995, Pattern Recognit..