A KNOWLEDGE-BASED APPROACH FOR SCRIPT RECOGNITION WITHOUT TRAINING

Authors
RAO PVS
Citation
Pvs. Rao, A KNOWLEDGE-BASED APPROACH FOR SCRIPT RECOGNITION WITHOUT TRAINING, IEEE transactions on pattern analysis and machine intelligence, 17(12), 1995, pp. 1233-1239
Citations number
17
Categorie Soggetti
Computer Sciences","Computer Science Artificial Intelligence","Engineering, Eletrical & Electronic
Journal title
IEEE transactions on pattern analysis and machine intelligence → ACNP
ISSN journal
01628828
Volume
17
Issue
12
Year of publication
1995
Pages
1233 - 1239
Database
ISI
SICI code
0162-8828(1995)17:12<1233:AKAFSR>2.0.ZU;2-F
Abstract
The approach is based on an empirical parametric model for the writing hand system. The parameters are so chosen and quantized as to retain only broad shape information ignoring writer-dependent and other varia bility. Concatenation of character prototypes generates archetypal ref erence words for recognition; training is unnecessary. Recognition sco res exceed 90%.