Title :
Hypercomplex Quality Assessment of Multi/Hyperspectral Images
Author :
Garzelli, Andrea ; Nencini, Filippo
Author_Institution :
Dept. of Inf. Eng., Univ. of Siena, Siena, Italy
Abstract :
This letter presents a novel image quality index which extends the Universal Image Quality Index for monochrome images to multispectral and hyperspectral images through hypercomplex numbers. The proposed index is based on the computation of the hypercomplex correlation coefficient between the reference and tested images, which jointly measures spectral and spatial distortions. Experimental results, both from true and simulated images, are presented on spaceborne and airborne visible/infrared images. The results prove accurate measurements of inter- and intraband distortions even when anomalous pixel values are concentrated on few bands.
Keywords :
distortion; geophysical techniques; remote sensing; airborne imagery; anomalous pixel values; hypercomplex correlation coefficient; hypercomplex number; hyperspectral images; interband distortion; intraband distortion; monochrome images; multispectral images; spaceborne imagery; universal image quality index; Hypercomplex correlation coefficient (CC); hypercomplex number; image quality assessment; spectral distortion;
Journal_Title :
Geoscience and Remote Sensing Letters, IEEE
DOI :
10.1109/LGRS.2009.2022650