Change detection for hyperspectral sensing in a transformed low-dimensional space
- Los Alamos National Laboratory
We present an approach to the problem of change in hyperspectral imagery that operates in a two-dimensional space. The coordinates in the space are related to Mahalanobis distances for the combined ('stacked') data and the individual hyperspectral scenes. Although it is only two-dimensional, this space is rich enough to include several well-known change detection algorithms, including the hyperbolic anomalous change detector, based on Gaussian scene clutter, and the EC-uncorrelated detector based on heavy-tailed (elliptically contoured) clutter. Because this space is only two-dimensional, adaptive machine learning methods can produce new change detectors without being stymied by the curse of dimensionality. We investigate, in particular, the utility of the support vector machine for learning boundaries in this 2-D space, and compare the performance of the resulting nonlinearly adaptjve detector to change detectors that have themselves shown good performance.
- Research Organization:
- Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- AC52-06NA25396
- OSTI ID:
- 984854
- Report Number(s):
- LA-UR-10-00455; LA-UR-10-455; TRN: US201016%%1642
- Resource Relation:
- Conference: Military Sensing Symposium on Battlefield Survivability & Discrimination ; February 22, 2010 ; Orlando, FL
- Country of Publication:
- United States
- Language:
- English
Similar Records
Hyperspectral target detection using manifold learning and multiple target spectra
VEGETATION COVER ANALYSIS OF HAZARDOUS WASTE SITES IN UTAH AND ARIZONA USING HYPERSPECTRAL REMOTE SENSING