Spatial-Spectral Schroedinger Eigenmaps

Graph-based dimensionality reduction technique for image data

Performs dimensionality reduction and classification of hyperspectral imagery using the Spatial-Spectral Schroedinger Eigenmaps (SSSE) algorithm, as described in the papers:
1) N. D. Cahill, W. Czaja, and D. W. Messinger, "Schroedinger Eigenmaps with Nondiagonal Potentials for Spatial-Spectral Clustering of Hyperspectral Imagery," Proc. SPIE Defense & Security: Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XX, May 2014.

2) N. D. Cahill, W. Czaja, and D. W. Messinger, "Spatial-Spectral Schroedinger Eigenmaps for Dimensionality Reduction and Classification of Hyperspectral Imagery," submitted.

This example script also performs classification using Support Vector Machines, as described in paper 2.