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

A radial basis function neural network (RBFNN) is developed to examine two mixing models, linear and nonlinear spectral mixtures, which describe the s pectra collected by both airborne and laboratory-based spectrometers. We ex amine the possibility that there may be naturally occurring situations wher e the typically used linear model may not provide the most accurate resulta nt spectral description. Under such a circumstance, a nonlinear model may b etter describe the mixing mechanism.