A quantitative and comparative analysis of linear and nonlinear spectral mixture models using radial basis function neural networks

Author

Guilfoyle, Kerri J. ; Althouse, Mark L. ; Chang, Chein-I

Author_Institution

Dept. of Comput. Sci. & Electr. Eng., Maryland Univ., Baltimore, MD, USA

Volume

39

Issue

10

fYear

2001

fDate

10/1/2001 12:00:00 AM

Firstpage

2314

Lastpage

2318

Abstract

A radial basis function neural network (RBFNN) is developed to examine two mixing models, linear and nonlinear spectral mixtures, which describe the spectra collected by both airborne and laboratory-based spectrometers. The authors examine the possibility that there may be naturally occurring situations where the typically used linear model may not provide the most accurate resultant spectral description. Under such a circumstance, a nonlinear model may better describe the mixing mechanism

Keywords

geophysical signal processing; geophysical techniques; geophysics computing; multidimensional signal processing; radial basis function networks; remote sensing; terrain mapping; comparative analysis; geophysical measurement technique; hyperspectral remote sensing; land surface; linear spectral mixture model; mixing models; multispectral remote sensing; neural net; nonlinear spectral mixture model; optical method; quantitative analysis; radial basis function neural network; terrain mapping; Hyperspectral imaging; Instruments; Laboratories; Optical imaging; Pixel; Radial basis function networks; Reflectivity; Rivers; Soil; Spectroscopy;

fLanguage

English

Journal_Title

Geoscience and Remote Sensing, IEEE Transactions on

Publisher

ieee

ISSN

0196-2892

Type

jour

DOI

10.1109/36.957296

Filename

957296