FEATURE-BASED NEURAL WAVELET OPTICAL CHARACTER-RECOGNITION SYSTEM

Authors
IFTEKHARUDDIN KM SCHECHINGER TD JEMILI K KARIM MA
Citation
Km. Iftekharuddin et al., FEATURE-BASED NEURAL WAVELET OPTICAL CHARACTER-RECOGNITION SYSTEM, Optical engineering, 34(11), 1995, pp. 3193-3199
Citations number
23
Categorie Soggetti
Optics
Journal title
Optical engineering → ACNP
ISSN journal
00913286
Volume
34
Issue
11
Year of publication
1995
Pages
3193 - 3199
Database
ISI
SICI code
0091-3286(1995)34:11<3193:FNWOCS>2.0.ZU;2-O
Abstract
A hybrid character recognition system that uses a feature-extraction m ethod is proposed. The features are extracted using a wavelet transfor m, preclassified using a k-nearest-neighbor-based neural net and subse quently postprocessed using an optical correlator. This feature-based neural wavelet optical architecture is then tested on blurred characte r images.