What question did this study set out to answer?

To develop an orthogonal subspace projection technique that simultaneously suppresses background interference, classifies target materials, and reduces data dimensionality in hyperspectral images.

July 1, 1994IEEE Transactions on Geoscience and Remote Sensing

Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach

View Full Paper

Ask AI

Bookmark

Authors

JHJoseph C. HarsanyiSeabrook

Discussion

Member takes

Implication

Algorithm development study demonstrates simultaneous interference suppression and target classification in hyperspectral imagery, indicating efficient dimensionality reduction.

Key Points

To develop an orthogonal subspace projection technique that simultaneously suppresses background interference, classifies target materials, and reduces data dimensionality in hyperspectral images.
Formulated an orthogonal subspace projection operator that projects pixel vectors onto a subspace orthogonal to undesired spectral signatures using least-squares interference suppression.
Projected residual signals onto target spectral signatures to maximize the signal-to-noise ratio for individual component images.
Extended the projection model to accommodate k distinct signatures of interest across both pure and spectrally mixed pixels.
Optimally eliminates interfering background signatures in the least-squares sense to produce high-contrast, classification-specific output images.
Reduces hyperspectral data dimensionality to k dimensions corresponding directly to the number of target signatures of interest.

Cite This Study

Joseph C. Harsanyi (1994) studied this question.

synapsesocial.com/papers/6a7d1676ade38b7dc346ccc7 https://doi.org/10.1109/36.298007

View Full Paper

Ask AI

Bookmark

Implication

Algorithm development study demonstrates simultaneous interference suppression and target classification in hyperspectral imagery, indicating efficient dimensionality reduction.

Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach

Discussion

Member takes

Implication

Key Points

Cite This Study

Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider