Share Email Print

Proceedings Paper

Exploitation of hyperspectral imagery using adaptive resonance networks
Author(s): Robert S. Rand
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Hyperspectral imagery consists of a large number of spectral bands that is typically modeled in a high dimensional spectral space by exploitation algorithms. This high dimensional space usually causes no inherent problems with simple classification methods that use Euclidean distance or spectral angle for a metric of class separability. However, classification methods that use quadratic metrics of separability, such as Mahalanobis distance, in high dimensional space are often unstable, and often require dimension reduction methods to be effective. Methods that use supervised neural networks or manifold learning methods are often very slow to train. Implementations of Adaptive Resonance Theory, such as fuzzy ARTMAP and distributed ARTMAP have been successfully applied to single band imagery, multispectral imagery, and other various low dimensional data sets. They also appear to converge quickly during training. This effort investigates the behavior of ARTMAP methods on high dimensional hyperspectral imagery without resorting to dimension reduction. Realistic-sized scenes are used and the analysis is supported by ground truth knowledge of the scenes. ARTMAP methods are compared to a back-propagation neural network, as well as simpler Euclidean distance and spectral angle methods.

Paper Details

Date Published: 24 September 2007
PDF: 12 pages
Proc. SPIE 6696, Applications of Digital Image Processing XXX, 66960M (24 September 2007); doi: 10.1117/12.734918
Show Author Affiliations
Robert S. Rand, U.S. Dept. of Defense (United States)

Published in SPIE Proceedings Vol. 6696:
Applications of Digital Image Processing XXX
Andrew G. Tescher, Editor(s)

© SPIE. Terms of Use
Back to Top