Digital image processing with dynamical neural networks for resource management: theories aspects
Luis Javier Morales-Mendoza, Oscar Gerardo Ibarra-Manzano, Mario-Alberto Ibarra-Manzano, Yuriy S. Shmaliy · 2006
In this paper, we present the theories aspects of the problem from the reconstruction and enhancing of radar imaging in the natural resource management with detection of specials characteristics. The problem is oriented to the data massive processing involving with recurrent neural networks. The Maximum Entropy Variational Analysis method is implemented into the recurrent neural networks modified Hopfield-type to the reconstruction and, detection and stopping edges (enhancing) in the radar imaging. Furthermore, here we present two forms of the computational implementation of the recurrent neural network.