EVALUATING THE TOPOLOGY-PRESERVATION CAPABILITIES OF A SELF-ORGANIZING LOGICAL NEURAL-NETWORK

Authors
TAMBOURATZIS G STONHAM TJ
Citation
G. Tambouratzis et Tj. Stonham, EVALUATING THE TOPOLOGY-PRESERVATION CAPABILITIES OF A SELF-ORGANIZING LOGICAL NEURAL-NETWORK, Pattern recognition letters, 14(11), 1993, pp. 927-934
Citations number
9
Categorie Soggetti
Computer Sciences, Special Topics","Computer Applications & Cybernetics
Journal title
Pattern recognition letters → ACNP
ISSN journal
01678655
Volume
14
Issue
11
Year of publication
1993
Pages
927 - 934
Database
ISI
SICI code
0167-8655(1993)14:11<927:ETTCOA>2.0.ZU;2-P
Abstract
An unsupervised learning algorithm which enables a logical neural netw ork to separate different classes of binary images while at the same t ime creating a topology-preserving mapping of the input space is descr ibed. Results concerning its successful application to character separ ation and recognition are presented and the quality of the mappings ge nerated by the system is evaluated.