Training Neural Networks for Reading Handwritten Amounts on Checks

8 Pages Posted: 1 Jun 2002  

Rafael Palacios

Universidad Pontificia Comillas

Amar Gupta

Massachusetts Institute of Technology (MIT)

Date Written: May 2002

Abstract

While reading handwritten text accurately is a difficult task for computers, the conversion of handwritten papers into digital format is necessary for automatic processing. Since most bank checks are handwritten, the number of checks is very high, and manual processing involves significant expenses, many banks are interested in systems that can read check automatically. This paper presents several approaches to improve the accuracy of neural networks used to read unconstrained numerals in the courtesy amount field of bank checks.

Keywords: Optical Character Recognition, Neural Networks, Document Imaging, Check Processing, Unconstrained Handwritten Numerals

Suggested Citation

Palacios, Rafael and Gupta, Amar, Training Neural Networks for Reading Handwritten Amounts on Checks (May 2002). MIT Sloan Working Paper No. 4365-02; Eller College Working Paper No. 1022-05. Available at SSRN: https://ssrn.com/abstract=314779 or http://dx.doi.org/10.2139/ssrn.314779

Rafael Palacios (Contact Author)

Universidad Pontificia Comillas ( email )

Alberto Aguilera 21
Madrid, Madrid 28015
Spain

Amar Gupta

Massachusetts Institute of Technology (MIT) ( email )

77 Massachusetts Avenue
Building 32-256
Cambridge, MA 02139
United States
617-253-0418 (Phone)

Paper statistics

Downloads
311
Rank
78,039
Abstract Views
3,851