Premium
A CNN handwritten character recognizer
Author(s) -
Suzuki H.,
Matsumoto T.,
Chua Leon O.
Publication year - 1992
Publication title -
international journal of circuit theory and applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.364
H-Index - 52
eISSN - 1097-007X
pISSN - 0098-9886
DOI - 10.1002/cta.4490200513
Subject(s) - numeral system , computer science , pattern recognition (psychology) , character (mathematics) , character recognition , artificial intelligence , speech recognition , feature (linguistics) , classifier (uml) , base (topology) , mathematics , mathematical analysis , linguistics , philosophy , geometry , image (mathematics)
CNNs are used for feature detection in handwritten character recognition. Detected features are fed to a simple classifier network. Performance was tested by using two well‐known ETL data base series: (i) ETL3 consisting of numerals, alphabets and several symbols and (ii) ETL8B2 consisting of Japanese Hirakana characters. the average recognition rate for ETL3 is 94.8%, while that for ETL8B2 is 85.7%. Both series include ‘hard’ characters so distorted that even humans cannot recognize them.