Spectral unmixing using nonnegative basis learning: comparison of geometrical and statistical endmember extraction algorithms

Mario Parente

doi:10.1117/12.777895

15 April 2008 Spectral unmixing using nonnegative basis learning: comparison of geometrical and statistical endmember extraction algorithms

Mario Parente

Author Affiliations +

Proceedings Volume 6960, Space Exploration Technologies; 69600P (2008) https://doi.org/10.1117/12.777895
Event: SPIE Defense and Security Symposium, 2008, Orlando, Florida, United States

Abstract

Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) data can be used to identify the presence of minerals on the surface or Mars. The data are a peculiar set from which to extract endmembers. Using an image from a previously investigate area of the surface, we compare a geometrical and a statistical algorithm for extracting endmembers for mineral identification. Both algorithms correctly identified the spectra of the two minerals known to be present in the Nili Fossae region of Mars. Both algorithms suffer from linearity assumption. Even though the statistical algorithm is less robust with respect to outliers, it has potential to extract endmembers in complex data clouds because of its local nature.

Citation Download Citation

Mario Parente "Spectral unmixing using nonnegative basis learning: comparison of geometrical and statistical endmember extraction algorithms", Proc. SPIE 6960, Space Exploration Technologies, 69600P (15 April 2008); https://doi.org/10.1117/12.777895

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available