Hyperspectral imaging of the earth: Remote analytical chemistry in an uncontrolled environment

Abstract

Imaging spectrometry, also called hyperspectral imaging, is gaining widespread interest as a remote sensing technique that allows for quantitative determination of the abundance and composition of Earth surface materials at the subpixel resolution level. Hyperspectral imaging sensors collect radiance data from airborne and spaceborne platforms that must be converted to apparent surface reflectance before analysis techniques can be brought to bear. Atmospheric correction techniques have been developed that use the data themselves to remove spectral atmospheric transmission and scattered path radiance. The reflectance data can be analyzed by both deterministic and statistical techniques. The statistical techniques include the use of the data themselves to determine the pure endmember spectra even though they do not exist as pure pixels in the image. Applications include mineral mapping and remote determination of forest canopy chemistry that is used in mapping species as well as determining C/N ratios. © 1996 John Wiley & Sons, Inc. Field Analyt Chem Technol 1: 67–76, 1996.