RELAXATION BY HOPFIELD NETWORK IN STEREO IMAGE MATCHING

Pajares, Gonzalo, Cruz García, Jesús Manuel de la y Aranda Almansa, Joaquín . (1998) RELAXATION BY HOPFIELD NETWORK IN STEREO IMAGE MATCHING. PPattern Recognition Letters (v.31, nº 5), p. 561-574


Título RELAXATION BY HOPFIELD NETWORK IN STEREO IMAGE MATCHING
Autor(es) Pajares, Gonzalo
Cruz García, Jesús Manuel de la
Aranda Almansa, Joaquín
Materia(s) Informática
Palabras clave Stereo correspondence
Smoothness
Similarity
HopÞeld neural network
Relaxation
Uniqueness
Editor(es) Elsevier
Fecha 1998-01-01
Formato application/pdf
Identificador PCA98b
bibliuned:677
Publicado en la Revista PPattern Recognition Letters (v.31, nº 5), p. 561-574
Idioma eng
Versión de la publicación publishedVersion
Tipo de recurso Article
Derechos de acceso y licencia http://creativecommons.org/licenses/by-nc-nd/4.0
info:eu-repo/semantics/openAccess
Tipo de acceso Acceso abierto
Notas adicionales ÐThis paper outlines a relaxation approach using the HopÞeld neural network for solving the global stereovision matching problem. The primitives used are edge segments. The similarity, smoothness and uniqueness constraints are transformed into the form of an energy function whose minimum value corresponds to the best solution of the problem. We combine two methods: (a) optimization/relaxation(1) and (b) relaxation merit(2) with the above three constraints mapped in an energy function. The main contribution is made (1) by applying a learning strategy in the similarity constraint and (2) by introducing speciÞc conditions to overcome the violation of the smoothness constraint and to avoid the serious problem arising from the required Þxation of a disparity limit. So, we improve the stereovision matching process. A better performance of the proposed method is illustrated with a comparative analysis against a classical relaxation method.

 
Versiones
Versión Tipo de filtro
Contador de citas: Google Scholar Search Google Scholar
Estadísticas de acceso: 220 Visitas  -  Estadísticas en detalle
Creado: Fri, 16 Nov 2007, 15:36:48 CET