AN EVALUATION OF TECHNIQUES FOR THE EXTRACTION OF MINERAL ABSORPTION FEATURES FROM HIGH SPECTRAL RESOLUTION REMOTE-SENSING DATA

Airborne imaging spectrometer data are influenced by a number of external factors, which mask subtle absorption features that permit the identification of surface mineralogy. This paper examines a variety of techniques developed to remove those factors, which result from the solar irradiance drop off, atmospheric absorption, and topographic effects. The techniques investigated are the flat-field correction, log residuals, and corrections using the LOWTRAN 7 atmospheric transfer code. These techniques were applied to Airborne Visible/InfraRed Imaging Spectrometer (AVIRIS) data acquired over Cuprite, Nevada. The processed data were evaluated for their ability to display the diagnostic absorption features of three areas of known mineralogy. These areas are dominated by the minerals alunite, buddingtonite, and kaolinite. The spectral features observed in the manipulated data were compared against those observed in the original data. Results indicate that the data corrected using the LOWTRAN 7 atmospheric transfer code constrained with local weather station data were the most effective at displaying the diagnostic absorption features of the areas of known mineralogy and introduced the least number of artifacts into the data. Of the remaining techniques, log residuals was the next most effective, based on the previous criteria, and has the additional advantage of not requiring any external data.