ANN-based handwritten character recognition

Author

Hanmin, Huang ; Xiyue, Huang ; Ping, Zhang ; Yi, Chai ; Weiren, Shi

Author_Institution

Inst. of Autom., Chongqing Univ., China

fYear

1999

fDate

36373

Firstpage

1177

Lastpage

1180

Abstract

Based on an artificial neural network, digital image processing, and features extraction theory, the authors analyzed a BP network´s defect then presented improving solutions. In this paper, a new kind of handwritten character system has been constructed. Referring to the shortcoming of the traditional BP algorithm, a modified learning factor with adaptation is introduced, and a bizarre sample feature database is constructed for speeding up modified BP learning and classification. Experimental results show that the modified BP neural network algorithm (three layers forward, no feedback) can be used in handwritten character recognition, and satisfactory results have been obtained

Keywords

backpropagation; feature extraction; feedforward neural nets; handwritten character recognition; multilayer perceptrons; ANN-based handwritten character recognition; adaptation; classification; digital image processing; modified BP neural network algorithm; modified learning factor; Automation; Biological neural networks; Brain; Character recognition; Cities and towns; Digital images; Fault tolerance; Feature extraction; Image analysis; Image databases;

fLanguage

English

Publisher

ieee

Conference_Titel

SICE Annual, 1999. 38th Annual Conference Proceedings of the

Conference_Location

Morioka

Print_ISBN

4-907764-13-8

Type

conf

DOI

10.1109/SICE.1999.788719

Filename

788719