| | |
Summary: MORPHOLOGICAL SCALE-SPACE FOR HYPERSPECTRAL IMAGES AND
DIMENSIONALITY EXPLORATION USING TENSOR MODELING
Santiago Velasco-Forero, Jes´us Angulo
Mines ParisTech, Center de Morphologie Math´ematique, Fontainebleau, France
ABSTRACT
This paper proposes a framework to integrate spatial infor-
mation into unsupervised feature extraction for hyperspectral
images. In this approach a nonlinear scale-space represen-
tation using morphological levelings is formulated. In or-
der to apply feature extraction, Tensor Principal Components
are computed involving spatial and spectral information. The
proposed method has shown significant gain over the conven-
tional schemes used with real hyperspectral images. In ad-
dition, the proposed framework opens a wide field for future
developments in which spatial information can be easily inte-
grated into the feature extraction stage. Examples using real
hyperspectral images with high spatial resolution showed ex-
cellent performance even with a low number of training sam-
ples.
Index Terms-- Unsupervised feature extraction, mathe-
|
