SIBGRAPI Digital Library Archive
1993 accepted paper update form
The submission system will allow any upgrade until December 27th, 2013.
If you have any question, please contact:
Don't forget to press the "+" button to enter the co-author data, if any.
Distance-discriminator neurons DDNs and their combination in Distance-discriminator Neural Networks DDNNs are proposed and discussed. DDNs, based on distance metric concepts, are able to discriminate whether a given point (x,y) belongs to a closed region such as diamond-, rectangle and ellipse-bound regions, which are tasks traditionally performed by perceptrons. DDNs can also be straightforwardly modified in order to discriminate hollow regions having as outer boundaries the above mentioned geometrical figures or even combinations of them. The principal advantage of DDNNs over perceptrons is a substantial reduction of execution time and/or the amount of required hardware operators: many polygonal classification regions which would otherwise demand large perceptron structures can be discriminated with only a few DDNNs. DDNNs can also be easily programmed by design or automatically with the help of the hough transform. Such issues as well as the relative advantages of DDNNs and perceptrons and a complete application example are presented and discussed in the present paper.
Síntese de Imagens
Reconhecimento de Padrões
Visualização de Dados
Visão por Computador
Aplicações em Medicina
Modelagem e Visualização
Animação e Multimídia
Processamento de Imagens
Previous Lower Units
:: Powered by UR