Please use this identifier to cite or link to this item:
Title: Object-based quality evaluation procedure for fused remote sensing imagery
Authors: Rodríguez-Esparragón, Dionisio 
Marcello, Javier 
Eugenio González, Francisco 
Garcia-Pedrero, Ángel
Gonzalo-Martin, Consuelo
UNESCO Clasification: 251007 Oceanografía física
250616 Teledetección (Geología)
120325 Diseño de sistemas sensores
Keywords: Image quality
Remote sensing
Human vision
Image object analysis
Issue Date: 2017
Journal: Neurocomputing 
Abstract: Satellite sensors usually provide two types of data: panchromatic and multispectral images which are characterized by their high spatial resolution and high spectral resolution respectively. In this context, the fusion techniques or pansharpening consist of merging these different aspects to obtain a fused (or pan-sharpened) image with high spatial and spectral resolutions. In this paper, a new quality assessment scheme for pan-sharpened remote sensing imagery is proposed. The methodology described extracts the segments of the images to constitute the basic elements of the measuring quality methodology. This new strategy overcomes traditional pixel-based perspectives, approaching an evaluation by human observers. The results of its application to a set of fused images show that an object-based assessment is consistent in terms of quality determination of both the spectral and spatial properties of remote sensing images.
ISSN: 0925-2312
DOI: 10.1016/j.neucom.2016.06.091
Source: Neurocomputing [ISSN 0925-2312], v. 255 , p. 40-51
Appears in Collections:Artículos

Show full item record
Recommend this item

Google ScholarTM




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