A novel continuous dual mode neural network in stereo-matching process

Abstract

In the present paper we describe completely innovative of architecture of artificial neural network based on Hopfield structure. It is analogue implementation of dual mode Hopfield-like network for solving stereo matching problem. Considered network consists of basic layer of neurons realized by analogue Hopfield-like network and managementing layer. Thanks to the using of managementing layer there is a possibility of modification of the connection weights between the neurons in basic layer. This enables of improvement and correction of the solution. In the present article we also describe energy function for basic layer and the condition of syntactic correctness of the solution, allowing to correct connection weights. The network considered here was taken under experimental tests using real stereo pictures as well simulated stereo images. © 2010 Springer-Verlag Berlin Heidelberg.