Automatic Target Recognition using a Constructive Approach

Ernesto Cuadros-VargasHumberto Costa de SousaAndré Carlos Ponce de Leon Ferreira de Carvalho

This paper presents the use of Constructive Neural Networks for Automatic Target Recognition. The work presented here is a contribution for the SAPRI project (System for Acquisition, Processing and Images Recognition) which is being developed for the Brazilian Navy. The objective of this paper is to investigate the behavior of Constructive Neural Networks in image recognition tasks, specially with images from the SAPRI project. Constructive Neural Networks differ from standard networks due to their ability to change their own number of elements by adding or removing units and connections. Three constructive algorithms are presented in this paper: Cascade Correlation, Tower and Pyramid.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: