International Joint Conference on Neural Networks 1989
DOI: 10.1109/ijcnn.1989.118578
|View full text |Cite
|
Sign up to set email alerts
|

A new approach for pattern recognition by neural networks with scramblers

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
2
0

Year Published

1990
1990
1993
1993

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 13 publications
(3 citation statements)
references
References 4 publications
0
2
0
Order By: Relevance
“…Preprocessing consisted of centering based on first-order moment information, and rough scaling down of large characters to fit within the input field of the network. As a result of this preprocessing, the recognition system could tolerate translation and scaling in the input data, but not specifically due to a neural solution to this variance as has been part of other algorithms [6,7].…”
Section: Methodsmentioning
confidence: 97%
“…Preprocessing consisted of centering based on first-order moment information, and rough scaling down of large characters to fit within the input field of the network. As a result of this preprocessing, the recognition system could tolerate translation and scaling in the input data, but not specifically due to a neural solution to this variance as has been part of other algorithms [6,7].…”
Section: Methodsmentioning
confidence: 97%
“…Analisys ofthe Dimensionality ofNeural Networks for Pattern Recognition. Paliem Recognition, v. 23, n. 10, p. 1131-1140 ,) Sistema de Reconhecimento de Padrões VhUAis Invariante a Trarufonnações Geométricas Utilizando Redes Newais Artificiais de Múltiplas Camadas FUKUMI, M.;TAKEDA, F.;KOSAKA, T. (1992). Rotation-invariant neural pattern recognition system with application to coin recognition, IEEE Trans.…”
Section: Conclusõesmentioning
confidence: 99%
“…Algumas modificações foram introduzidas no modelo original do Neocognitron para tentar superar as deficiências apresentadas no mesmo (HOSOKAWA et al, 1989;CRUZ et al, 1989, FUKUMI et al, 1992. Tais modificações resultaram em melhor desempenho que o modelo original, sendo aplicadas em alguns tipos de problemas práticos de reconhecimento de objetos.…”
Section: Outras Técnicasunclassified