Skeletonization of sparse shapes using dynamic competitive neural networks

Mostrar todas las versiones(2)
Detalles Bibliográficos
Autor Principal: Hasperué, Waldo
Otros autores o Colaboradores: Corbalán, Leonardo César, Bria, Oscar Norberto, Lanzarini, Laura Cristina
Formato: Capítulo de libro
Lengua:inglés
Temas:
REDES NEURONALES
PROCESAMIENTO DE IMÁGENES
RECONOCIMIENTO DE PATRONES
Acceso en línea:http://goo.gl/lNoljp
Consultar en el Cátalogo
Resumen:The detection of regions and objects in digital images is a topic of utmost importance for solving several problems related to the area of pattern recognition. In this direction, skeletonization algorithms are a widely used tool since they allow us to reduce the quantity of available data, easing the detection of characteristics for their recognition and classification. In addition, this transformation of the original data in its essential characteristics eases the elimination of local noise which is present in the data input. This paper proposes a new skeletonization strategy applicable to sparse images from a competitive, dynamic neural network trained with the AVGSOM method. The strategy developed in this paper determines the arc making up the skeleton combining AVGSOM non-supervised learning with a minimum spanning tree. The proposed method has been applied in images with different spanning shape and degree. In particular, the results obtained have been compared to existing solutions, showing successful results. Finally, some conclusions, together with some future lines of work, are presented.
Notas:Formato de archivo: PDF. -- Este documento es producción intelectual de la Facultad de Informática - UNLP (Colección BIPA/Biblioteca)
Descripción Física:1 archivo (376,6 KB)

MARC

LEADER 00000naa a2200000 a 4500
003 AR-LpUFIB
005 20250423183150.0
008 230201s2006 xx o 000 0 eng d
024 8 |a DIF-M6581  |b 6720  |z DIF006001 
040 |a AR-LpUFIB  |b spa  |c AR-LpUFIB 
100 1 |a Hasperué, Waldo  |9 45531 
245 1 0 |a Skeletonization of sparse shapes using dynamic competitive neural networks 
300 |a 1 archivo (376,6 KB) 
500 |a Formato de archivo: PDF. -- Este documento es producción intelectual de la Facultad de Informática - UNLP (Colección BIPA/Biblioteca) 
520 |a The detection of regions and objects in digital images is a topic of utmost importance for solving several problems related to the area of pattern recognition. In this direction, skeletonization algorithms are a widely used tool since they allow us to reduce the quantity of available data, easing the detection of characteristics for their recognition and classification. In addition, this transformation of the original data in its essential characteristics eases the elimination of local noise which is present in the data input. This paper proposes a new skeletonization strategy applicable to sparse images from a competitive, dynamic neural network trained with the AVGSOM method. The strategy developed in this paper determines the arc making up the skeleton combining AVGSOM non-supervised learning with a minimum spanning tree. The proposed method has been applied in images with different spanning shape and degree. In particular, the results obtained have been compared to existing solutions, showing successful results. Finally, some conclusions, together with some future lines of work, are presented. 
534 |a Congreso Argentino de Ciencias de la Computación (12º : 2006 oct. 17-21 : Potrero de los Funes, San Luis), pp.1331-1341 
650 4 |a REDES NEURONALES  |9 42953 
650 4 |a PROCESAMIENTO DE IMÁGENES  |9 43134 
650 4 |a RECONOCIMIENTO DE PATRONES  |9 42951 
700 1 |a Corbalán, Leonardo César  |9 44782 
700 1 |a Bria, Oscar Norberto  |9 44747 
700 1 |a Lanzarini, Laura Cristina  |9 43377 
856 4 0 |u http://goo.gl/lNoljp 
942 |c CP 
952 |0 0  |1 0  |4 0  |6 A0339  |7 3  |8 BD  |9 80957  |a DIF  |b DIF  |d 2025-03-11  |l 0  |o A0339  |r 2025-03-11 17:04:19  |u http://catalogo.info.unlp.edu.ar/meran/getDocument.pl?id=606  |w 2025-03-11  |y CP 
999 |c 55783  |d 55783 

Ejemplares similares