Please note, due to essential maintenance online transactions will not be possible between 09:00 and 13:00 BST, on Monday 20th January 2020 (04:00-08:00 EDT). We apologise for any inconvenience.
Multi- and hyperspectral sensors in the visible to short-wave infrared (0.4–2.5 μm) are sensitive to spectral features caused by electronic charge transfer and transition metal crystal field band as well as molecular overtone absorptions. This chapter reviews several processing techniques used to map materials on planetary surfaces based on their reflectance spectra in this spectral region. Techniques that are reviewed include spectral matching in the form of spectral angle and spectral information divergence, linear and nonlinear spectral unmixing, partial unmixing/matched filters, and machine learning approaches in the form of self-organizing maps, neural network classification, and support vector machines.