Optical character recognition: a technology driver for neural networks

Author

Howard, R.E. ; Boser, B. ; Denker, J.S. ; Graf, H.P. ; Henderson, D. ; Hubbard, W. ; Jackel, L.D. ; Le Cun, Y. ; Baird, H.S.

Author_Institution

AT&T Bell Lab., Holmdel, NJ, USA

fYear

1990

fDate

1-3 May 1990

Firstpage

2433

Abstract

It is shown that a neural net can perform handwritten digit recognition with state-of-the-art accuracy. The solution required automatic learning and generalization from thousands of training examples and also required designing into the system considerable knowledge about the task-neither engineering nor learning from examples alone would have sufficed. The resulting network is well suited for implementation on workstations or PCs and can take advantage of digital signal processors (DSPs) or custom VLSI

Keywords

VLSI; digital signal processing chips; learning systems; neural nets; optical character recognition; automatic learning; custom VLSI; digital signal processors; handwritten digit recognition; neural networks; task; technology driver; training; workstations; Character recognition; Design engineering; Digital signal processing; Digital signal processors; Handwriting recognition; Knowledge engineering; Neural networks; Optical character recognition software; Personal communication networks; Workstations;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1990., IEEE International Symposium on

Conference_Location

New Orleans, LA

Type

conf

DOI

10.1109/ISCAS.1990.112502

Filename

112502