Methods for enhancing neural network handwritten character recognition

Author

Garris, M.D. ; Wilkinson, R.A. ; Wilson, C.L.

Author_Institution

Nat. Inst. of Stand. & Technol., Gaithersburg, MD, USA

Volume

i

fYear

1991

fDate

8-14 Jul 1991

Firstpage

695

Abstract

An efficient method for increasing the generalization capacity of neural character recognition is presented. The network uses a biologically inspired architecture for feature extraction and character classification. The numerical methods used are optimized for use on massively parallel array processors. The method for training set construction, when applied to handwritten digit recognition, yielded a writer-independent recognition rate of 92%. The activation strength produced by network recognition is an effective statistical confidence measure of the accuracy of recognition. A method of using the activation strength for reclassification is described which, when applied to handwritten digit recognition, reduced substitutional errors to 2.2%

Keywords

character recognition; computerised pattern recognition; neural nets; parallel algorithms; accuracy; activation strength; biologically inspired architecture; character classification; digit recognition; feature extraction; generalization capacity; handwritten character recognition; massively parallel array processors; neural network; numerical methods; reclassification; statistical confidence measure; substitutional errors; training set construction; writer-independent recognition rate; Bayesian methods; Character recognition; Error analysis; Feature extraction; Handwriting recognition; Image reconstruction; Least squares methods; NIST; Neural networks; Optimization methods;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on

Conference_Location

Seattle, WA

Print_ISBN

0-7803-0164-1

Type

conf

DOI

10.1109/IJCNN.1991.155265

Filename

155265