Rede morfológica não-supervisionada-RMNS / Unsupervisedmorphologicnet-RMNS
AUTOR(ES)
Francisco de Assis Tavares Ferreira Silva
DATA DE PUBLICAÇÃO
1998
RESUMO
This work proposes a new paradigm of Artificial Neural Net (RNA): The Unsupervised Morphologic Net (RMNS), characterized by unsupervised learning. This paradigm belongs to a class of translation invariant nets and it is based on of Mathematical Morphology (MM), Carpenter and Grossberg s ART, and Kohonen net models. At activation time, the template matching operator is implemented using translation invariant MM elementary operators. This operator propitiates a robust pattern detection with respect to addictive or subtractive noise, and/or for small rotations of the patterns to be recognized. In what it concerns to the training, the RMNS uses a Kohonen learning rule variation and a reset system inspired in the ART model proposed by Carpenter and Grossberg.
ASSUNTO(S)
morfologia matemática reconhecimento de padrões unsupervised learning aprendizagem de máquina machine learning inteligência artificial redes neurais aprendizagem não supervisionada artificial intelligence patern recognition mathematical morphology neural nets
