A2iA Check Reader: A Family of Bank Check Recognition Systems

Authors:
N. Gorski;V. Anisimov;E. Augustin;O. Baret;D. Price;J.-C. Simon
Affiliations:
-;-;-;-;-;-
Venue:
ICDAR '99 Proceedings of the Fifth International Conference on Document Analysis and Recognition
Year:
1999

Citing 0
Cited 8

Using a statistical language model to improve the performance of an HMM-based cursive handwriting recognition systems

Hidden Markov models
Optical Character Recognition for Cursive Handwriting

IEEE Transactions on Pattern Analysis and Machine Intelligence
Parsing N-Best Lists of Handwritten Sentences

ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 1
Offline Grammar-Based Recognition of Handwritten Sentences

IEEE Transactions on Pattern Analysis and Machine Intelligence
A system for processing handwritten bank checks automatically

Image and Vision Computing
Off-line handwritten word recognition using multi-stream hidden Markov models

Pattern Recognition Letters
Arabic bank check analysis and zone extraction

ICIAR'12 Proceedings of the 9th international conference on Image Analysis and Recognition - Volume Part I
Efficient Word Segmentation and Baseline Localization in Handwritten Documents Using Isothetic Covers

International Journal of Digital Library Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

The paper presents a new family of bank check recognition systems named A2iA Check Reader. Systems are designed to process omni-writer/omni-printer bank checks and associated payment documents written/printed in France, UK or US. All the systems have identical architecture and design principles, however, each of them contains a country-specific part and trained with a country-specific data. A2iA Check Reader has two recognition chains: one for courtesy- and legal amount recognition on handwritten documents, and the other - for printed documents amount recognition. The systems performs extraction and cleaning of amounts fields in document image; segmentation of text image into characters and words; recognition of characters and words; interpretation of amount and decision making to accept or reject the check. Recognition rate is 80-90%. In production mode, document acceptance rate is 60-75% with an error rate corresponding to that of a human operator.