Invariant character recognition in Dynamic Link Architecture

Author

Liu, James N K ; Lee, Raymond S T

Author_Institution

Dept. of Comput., Hong Kong Polytech. Univ., Hong Kong

fYear

1997

fDate

35738

Firstpage

188

Lastpage

195

Abstract

A model based on the application of the Dynamic Link Architecture (DLA) is presented for the off-line recognition of Chinese characters. It is a revised DLA model employing 4-vector dynamic link assignment instead of the original 3-vector links. A sample set of Chinese characters was used to test the performance of the model under various transformations including translation, reflection, rotation, dilation and distortion. Challenging results are obtained. An improvement of 40% in recognition rate was attained by using the revised DLA model for Chinese character recognition. On the other hand, for the testing of invariant properties, an overall correct recognition rate of 85% was obtained under various transformations

Keywords

brain models; character recognition; invariance; neural net architecture; 4-vector dynamic link assignment; Dynamic Link Architecture; dilation; distortion; invariant character recognition; long-term memory; neural architecture; neural networks; off-line Chinese character recognition; performance; recognition rate; reflection; rotation; short-term memory; temporary links; transformations; translation; Biological neural networks; Biological system modeling; Character recognition; Computer architecture; Handwriting recognition; Neural networks; Neurons; Reflection; System performance; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Knowledge and Data Engineering Exchange Workshop, 1997. Proceedings

Conference_Location

Newport Beach, CA

Print_ISBN

0-8186-8230-2

Type

conf

DOI

10.1109/KDEX.1997.629865

Filename

629865