Year: 1993
Type: article
Abstract: Spectral mixture analysis, or unmixing, has proven to be a useful tool in the semi-quantitative interpretation of AVIRIS data. Using a linear mixing model and a set of hypothesized endmember spectra, unmixing seeks to estimate the fractional abundance patterns of the various materials occurring within the imaged area. However, the validity and accuracy of the unmixing rest heavily on the 'user-supplied' set of endmember spectra. Current methods for emdmember determination are the weak link in the unmixing chain.
Author: Joseph W. Boardman
Institution: University of Colorado Boulder
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W1555549210