Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/74021
Title: Optic flow estimation in fluid images I
Authors: Alemán Flores, Miguel 
Álvarez, L 
Gonzalez, E. 
Mazorra, L. 
Sánchez, J. 
UNESCO Clasification: 220990 Tratamiento digital. Imágenes
120304 Inteligencia artificial
120601 Construcción de algoritmos
120602 Ecuaciones diferenciales
120326 Simulación
Issue Date: 2005
Project: Fluid Image Analysis And Description 
Journal: Cuadernos del Instituto Universitario de Ciencias y Tecnologías Cibernéticas 
Abstract: In this work we present some comparative qualitative results of optic flow estimation in PIV and satellite fluid image sequence. This work has been done in the context of the FLUID Specific Targeted Research project - Contract No 513633 founded by the CEE. The main goal of this paper is to compare in a very basic way 3 optic flow estimation methods in the context of some fluid image sequence provided to us by AEROBIO-CEMAGREF and the ”Laboratorire de M´et´orologie Dynamique de la Ecole Polytechnique” both partners of the FLUID project. In order to simplify the comparison and better understand the comparison results we have used just 2 images of each sequence, so no temporal smoothing or another 3D preprocessing have been performed. We have also implemented ourself in C language the 3 optic flow methods in order to better understand each method and to be sure about what are we doing. The organization of the paper is as follows: In section 1, we describe the optic flow methods we use for comparison purposes. In section 2 we present an invariant analysis of the methods. In section 3, we present the numerical experiments and finally in section 4 we present some conclusions and future works.
URI: http://hdl.handle.net/10553/74021
ISSN: 1575-6807
Source: Cuadernos del Instituto Universitario de Ciencias y Tecnologías Cibernéticas [ISSN 1575-6807], n. 0031, p. 1-25
Appears in Collections:Artículos
Thumbnail
PDF
Adobe PDF (19,36 MB)
Show full item record

Page view(s)

95
checked on Sep 13, 2024

Download(s)

62
checked on Sep 13, 2024

Google ScholarTM

Check


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.