A NOVEL FEATURE RECOGNITION NEURAL-NETWORK AND ITS APPLICATION TO CHARACTER-RECOGNITION

被引：36
作者：
HUSSAIN, B
KABUKA, MR
机构：
[1] Department of Electrical and Computer Engineering, University of Miami, Coral Gables, FL
来源：
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE | 1994年 / 16卷 / 01期
关键词：
D O I：
10.1109/34.273711
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
This paper presents a Feature Recognition Network for pattern recognition that learns the patterns by remembering their different segments. The base algorithm for this network is a Boolean net algorithm that we developed during our past research. Simulation results show that the network can recognize patterns after significant noise, deformation, translation and even scaling. The network is compared to existing popular networks used for the same purpose, especially Neocognitron. The network is also analyzed as regards to interconnection complexity and information storage/retrieval.
引用
收藏
页码：98 / 106
页数：9
相关论文
共 41 条
[31]  
MORASHITA I, 1972, CYBERN, V11, P154
[32]  
ROSENBLATT F, 1959, PRINCIPLES NEURODYNA
[33]  
Rumelhart DE, 1986, ENCY DATABASE SYST, P45
[34]  
SCHWARTZ T, 1989, SEP IEEE VID C NEUR
[35]  
SKLANSKY J, 1981, PATTERN CLASSIFIERS
[36]  
SUNG C, 1990, JUN INT JOINT NEUR N, V3, P753
[37]   A DIGITAL ARCHITECTURE EMPLOYING STOCHASTICISM FOR THE SIMULATION OF HOPFIELD NEURAL NETS [J].
VANDENBOUT, DE ;
MILLER, TK .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS, 1989, 36 (05) :732-738
[38]  
WERBOS P, 1988, P IEEE INT C NEURAL
[39]  
WERBOS P, 1987, IEEE T SYST MAN CYBE, V17
[40]  
WIDROW B, 1988, IEEE COMPUTER MAR, P25
12345