A new hybrid methodology for intelligent Chinese character recognition

AL-DABASS, D., EVANS, D. and REN, M., 2005. A new hybrid methodology for intelligent Chinese character recognition. In: Proceedings - HIS'04: 4th International Conference on Hybrid Intelligent Systems. Piscataway, NJ, US: Institute of Electrical and Electronics Engineers Computer Society, pp. 104-109. ISBN 0769522912

Full text not available from this repository.

Abstract

A new methodology is proposed to combine fuzzy possibilistic reasoning with knowledge mining of syntax dynamics using hybrid recurrent nets. The structure of Chinese characters consist of a 3-layer hierarchy of character, radical and stroke. Fuzzy possibilistic reasoning is an appropriate set of algorithmic tools to aid automatic recognition of these characters. Associative memory artificial neural network algorithms form a suitable technique for realising these concepts. Several issues are explored: vagueness of radicals, their situation, position invariance, extraction order and shape. Hybrid recurrent nets are proposed to deal with recognition at the syntax level. Each character is the 2-D output of a syntax generating system which is subjected to a knowledge mining process to determine its behaviour parameters. The knowledge mining architecture consists of an extensible recurrent hybrid net hierarchy of multi-agents where the composite behaviour of agents at any one level is determined by those of the level is determined by those of the level immediately below.

Item Type: Chapter in book
Creators: Al-Dabass, D., Evans, D. and Ren, M.
Publisher: Institute of Electrical and Electronics Engineers Computer Society
Place of Publication: Piscataway, NJ, US
Date: 2005
Divisions: Schools > School of Science and Technology
Depositing User: EPrints Services
Date Added: 09 Oct 2015 10:03
Last Modified: 19 Oct 2015 14:26
URI: http://irep.ntu.ac.uk/id/eprint/7030

Actions (login required)

Edit View Edit View

Views

Views per month over past year

Downloads

Downloads per month over past year