Spectroscopic Analysis for Material Identification and Mapping Using Prism, an Envi/idl Based Software Package

The identification of materials by measuring and analyzing their reflectance spectra has been an important method used in analytical chemistry for decades. Now, airborne and space-based imaging spectrometers give scientists the opportunity to detect materials and map their distributions across the landscape. With new satelliteborne sensor planned for the future, for example HYSPIRI (HYPerspectral InfraRed Imager), robust methods are needed to fully exploit the information content in hyperspectral remote sensing data. Careful processing of laboratory, field, and remote sensing data is required to produce high quality spectra of materials of unknown composition which can be compared with spectral standards for identification. In this paper, an ENVI/IDL (ENvironment for Visualizing Images/Interactive Data Language) based software package called PRISM (Processing Routines in IDL for Spectroscopic Measurements) is described. Among its many functions, PRISM contains routines for the storage of spectra in database files, importation of field spectra, correction of field spectra to absolute reflectance, arithmetic operations on spectra, interactive continuum removal and comparison of absorption features, calibration of imaging spectrometer data to reflectance using field-measured spectra, identification and mapping of materials using spectral feature based analysis of reflectance data in an expertsystem framework, and import/export of ENVI spectral library files. PRISM has been used to calibrate image cubes and map materials for the full-country coverage of HyMap data collected in 2007 over the country of Afghanistan. The PRISM software will be released to the general public by the U.S. Geological Survey (USGS).