Structured Gaussian Components for Hyperspectral Image Classification

Berge, A.; Solberg, Anne

doi:10.1109/tgrs.2006.880626

Cited by 43 publications

(29 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A GMM [8] can be viewed as a combination of two or more normal Gaussian distributions. In a typical GMM representation, a probability density function for X = {x i } n i=1 in R d is written as the sum of K Gaussian components (modes); i.e.,…”

Section: Gmmmentioning

confidence: 99%

“…Many recent studies have demonstrated that spatial information is helpful for hyperspectral image classification since spatial texture is useful for hyperspectral images, especially at high resolution. In this work, we investigate LFDA as well as LPP coupled with a Gaussian-mixture-model (GMM) [8] classifier to improve the classification performance based on spatial-spectral information. The combination of LPP and the GMM classifier can be effective for hyperspectral image classification.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Locality-Preserving Discriminant Analysis in Kernel-Induced Feature Spaces for Hyperspectral Image Classification

Prasad

Fowler

et al. 2011

IEEE Geosci. Remote Sensing Lett.

111

View full text Add to dashboard Cite

Locality-preserving projection as well as local Fisher discriminant analysis is applied for dimensionality reduction of hyperspectral imagery based on both spatial and spectral information. These techniques preserve the local geometric structure of hyperspectral data into a low-dimensional subspace wherein a Gaussian-mixture-model classifier is then considered. In the proposed classification system, local spatial information-which is expected to be more multimodal than strictly spectral features-is used. Results with experimental hyperspectral data demonstrate that this system outperforms traditional classification approaches.

show abstract

Section: Gmmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Locality-Preserving Discriminant Analysis in Kernel-Induced Feature Spaces for Hyperspectral Image Classification

Prasad

Fowler

et al. 2011

IEEE Geosci. Remote Sensing Lett.

111

View full text Add to dashboard Cite

show abstract

“…In a previous study [13], we considered the idea of regularization by sharing covariance structure, as an extension of model-based clustering [14]. The structure sharing was obtained by sharing characteristics of the covariance eigendecomposition among all clusters in the data, allowing clusters to share orientation, shape, and volume.…”

Section: Related Workmentioning

confidence: 99%

Sparse Inverse Covariance Estimates for Hyperspectral Image Classification

Berge

Jensen

Solberg

2007

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

Abstract-Classification of remotely sensed hyperspectral images calls for a classifier that gracefully handles high-dimensional data, where the amount of samples available for training might be very low relative to the dimension. Even when using simple parametric classifiers such as the Gaussian maximum-likelihood rule, the large number of bands leads to copious amounts of parameters to estimate. Most of these parameters are measures of correlations between features. The covariance structure of a multivariate normal population can be simplified by setting elements of the inverse covariance matrix to zero. Well-known results from time series analysis relates the estimation of the inverse covariance matrix to a sequence of regressions by using the Cholesky decomposition. We observe that discriminant analysis can be performed without inverting the covariance matrix. We propose defining a sparsity pattern on the lower triangular matrix resulting from the Cholesky decomposition, and develop a simple search algorithm for choosing this sparsity. The resulting classifier is used on four different hyperspectral images, and compared with conventional approaches such as support vector machines, with encouraging results.

show abstract

“…The number of mixture components, M , is also determined using crossvalidation, see. 11 Parameter estimates of λ k , A k and D k are found in.…”

Section: Structured Eigenvector Decomposition Of the Covariance Matrixmentioning

confidence: 99%

Robust classification of hyperspectral images

Berg

Jensen

2007

SPIE Proceedings

View full text Add to dashboard Cite

This paper discusses robust classification of hyperspectral images. Both methods for dimensionality reduction and robust estimation of classifier parameters in full dimension are presented. A new approach to dimensionality reduction that uses piecewise constant function approximation of the spectral curve is compared to conventional dimensionality reduction methods like principal components, feature selection, and decision boundary feature extraction. Computing robust estimates of the decision boundary in full dimension is an alternative to dimensionality reduction. Two recently proposed techniques for covariance estimation based on the eigenvector decomposition and the Cholesky decomposition are compared to Support Vector Machine classifiers, simple regularized estimates, and regular quadratic classifiers. The experimental results on four different hyperspectral data sets demonstrate the importance of using simple, sparse models. The sparse model using Cholesky decomposition in full dimension performed slightly better than dimensionality reduction. However, if speed is an issue, the piecewise constant function approximation method for dimensionality reduction could be used.

show abstract

Structured Gaussian Components for Hyperspectral Image Classification

Cited by 43 publications

References 25 publications

Locality-Preserving Discriminant Analysis in Kernel-Induced Feature Spaces for Hyperspectral Image Classification

Locality-Preserving Discriminant Analysis in Kernel-Induced Feature Spaces for Hyperspectral Image Classification

Sparse Inverse Covariance Estimates for Hyperspectral Image Classification

Robust classification of hyperspectral images

Contact Info

Product

Resources

About