DocumentCode :
786397
Title :
Coded block neural network VLSI system using an adaptive learning-rate technique to train Chinese character patterns
Author :
Chen, Brian Y. ; Kuo, J.B.
Author_Institution :
Nat. Taiwan Univ., Taipei, Taiwan
Volume :
28
Issue :
21
fYear :
1992
Firstpage :
1941
Lastpage :
1942
Abstract :
A coded block neural network VLSI system is presented that uses an adaptive learning-rate technique to train Chinese character patterns. Using the adaptive learning-rate technique, 500 Chinese characters have been successfully trained in 47.2 h using a 28 MIPs computer.
Keywords :
VLSI; character recognition; image recognition; learning systems; neural nets; Chinese character pattern recognition; adaptive epoch based backpropagation; adaptive learning-rate technique; coded block neural network VLSI system;
fLanguage :
English
Journal_Title :
Electronics Letters
Publisher :
iet
ISSN :
0013-5194
Type :
jour
DOI :
10.1049/el:19921244
Filename :
170848
Link To Document :
بازگشت