Pattern recognition through optimization: experiments with spectrometer data

Multispectral imagery plays a significant role in Earth resource survey and evaluation and has been an essential part of terrestrial and planetary exploration. The capabilities of instruments in resolution and spectra discrimination are constantly being improved upon to meet the increasingly expanded requirements in many fields of research. With the increasing utilization of imaging spectrometer data, automatic identification of spectral signatures emanating from this imagery would be an invaluable facility as a precursor to classifying each pixel. Existing methods for identifying constituent spectra typically rely on spectra that are selected either manually or involve manual intervention. The aim of our research work is to assess the viability of linear and nonlinear approaches for spectral recognition and to devise techniques suitable for fully-automatic analysis. The techniques considered are numerical optimization, genetic algorithms, artificial neural networks, singular value decomposition (SVD), and a technique that uses SVD in an iterative scheme to avoid over-fitting. The ability of these methods to distinguish between a large number (up to 160) of different spectra is assessed, as is their stability in the presence of noise and their capacity to identify correctly combinations of spectra