DMCA
Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach (1994)
| Venue: | IEEE Transactions on Geoscience and Remote Sensing |
| Citations: | 187 - 16 self |
BibTeX
@ARTICLE{Harsanyi94hyperspectralimage,
author = {Joseph C. Harsanyi and Chein-i Chang and Senior Member},
title = {Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
year = {1994},
volume = {32},
pages = {779--785}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstruct-Most applications of hyperspectral imagery re-quire processing techniques which achieve two fundamental goals: 1) detect and classify the constituent materials for each pixel in the scene; 2) reduce the data volumeldimensionality, without loss of critical information, so that it can be processed efficiently and assimilated by a human analyst. In this paper, we describe a technique which simultaneously reduces the data dimensionality, suppresses undesired or inter-fering spectral signatures, and detects the presence of a spec-tral signature of interest. The basic concept is to project each pixel vector onto a subspace which is orthogonal to the unde-sired signatures. This operation is an optimal interference suppression process in the least squares sense. Once the inter-fering signatures have been nulled, projecting the residual onto the signature of interest maximizes the signal-to-noise ratio and results in a single component image that represents a classifi-cation for the signature of interest. The orthogonal subspace projection (OSP) operator can be extended to k signatures of interest, thus reducing the dimensionality of k and classifying the hyperspectral image simultaneously. The approach is ap-plicable to both spectrally pure as well as mixed pixels. I.
Keyphrases
dimensionality reduction hyperspectral image classification orthogonal subspace projection approach pixel vector basic concept abstruct-most application fundamental goal data dimensionality single component image signal-to-noise ratio mixed pixel critical information inter-fering signature inter-fering spectral signature constituent material spec-tral signature unde-sired signature optimal interference suppression process square sense orthogonal subspace projection human analyst hyperspectral image data volumeldimensionality hyperspectral imagery re-quire processing technique
