Supervised band selection for optimal use of data from airborne hyperspectral sensors

Riedmann, M.; Milton, E.J.

doi:10.1109/igarss.2003.1294245

Cited by 23 publications

(10 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[29][30][31][32][33][34] In a paper by Paskaleva and Hayat, 35 the performance of optimized band selection is compared directly to the performance of the MTI bands. Shen and Bassett 36 addressed this problem using an information theoretic approach.…”

Section: Designing Multispectral Imagersmentioning

confidence: 99%

Sparse linear filters for detection and classification in hyperspectral imagery

Theiler

Glocer

2006

SPIE Proceedings

View full text Add to dashboard Cite

We investigate the use of convex optimization to identify sparse linear filters in hyperspectral imagery. A linear filter is sparse if a large fraction of its coefficients are zero. A sparse linear filter can be advantageous because it only needs to access a subset of the available spectral channels, and it can be applied to high-dimensional data more cheaply than a standard linear detector. Finding good sparse filters is nontrivial because there is a combinatorially large number of discrete possibilities from which to choose the optimal subset of nonzero coefficients. But, by converting the optimality criterion into a convex loss function, and by employing an L1 penalty, one can obtain sparse solutions that are globally optimal. We investigate the performance of these sparse filters as a function of their sparsity, and compare the convex optimization approach with more traditional alternatives for feature selection. The methodology is applied both to the adaptive matched filter for weak signal detection, and to the Fisher linear discriminant for terrain categorization.

show abstract

Section: Designing Multispectral Imagersmentioning

confidence: 99%

Sparse linear filters for detection and classification in hyperspectral imagery

Theiler

Glocer

2006

SPIE Proceedings

View full text Add to dashboard Cite

show abstract

“…The existing band selection algorithms can be divided into two groups, namely the supervised and the unsupervised. The supervised band selection needs labeled training samples to train models, because of such prior information, supervised methods often have better performance when compared with the unsupervised ones [3], [4]. But the gathering of training samples may be too expensive or even impracticable [5], [6], this makes the supervised method hardly to be adopted for band selection.…”

Section: Introductionmentioning

confidence: 99%

Automatic Band Selection Using Spatial-Structure Information and Classifier-Based Clustering

Cao

Tao

et al. 2016

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Band selection plays an important role in hyperspectral image processing, which can reduce subsequent computation and storage requirement. There are two problems that are rarely investigated for band selection. First, some low-discriminating bands need to be manually removed by experts, which is time consuming and expensive; second, how to automatically determine the number of selected bands is not well investigated, though this is an indispensable step in practical applications. In this paper, we propose an automatic band selection (ABS) method to solve these problems. First, we exploit spatial structure to determine the discriminating power of each band, these bands with little structure information will be discarded; then, a powerful classifier is used for clustering, which can automatically find the underlying number of clusters. Experiments based on three real hyperspectral datasets demonstrate the effectiveness of our method.Index Terms-Automatic band selection (ABS), clustering with classifier, low-discriminating bands, number of bands.

show abstract

“…In [4], Archibald and Fann introduced an embedded feature selection algorithm that is tailored to operate with support vector machines (SVMs). In [5], bands were selected according to transformed divergence measure and the maximum likelihood accuracy on the test data.…”

Section: Introductionmentioning

confidence: 99%

Semisupervised Hyperspectral Band Selection Via Spectral–Spatial Hypergraph Model

Bai

Guo

Wang

et al. 2015

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Band selection is an essential step toward effective and efficient hyperspectral image classification. Traditional supervised band selection methods are often hindered by the problem of lacking enough training samples. To address this problem, we propose a semisupervised band selection method that allows contribution from both labeled and unlabeled hyperspectral pixels. This method first builds a hypergraph model from all hyperspectral samples to measure the similarity among pixels. We show that hypergraph can capture relationship among pixels in both spectral and spatial domain. In the second step, a semisupervised learning method is introduced to propagate class labels to unlabeled samples. Then a linear regression model with group sparsity constraint is used for band selection. Finally, hyperspectral pixels with selected bands are used to train a support vector machine (SVM) classifier. The proposed method is tested on three benchmark datasets. Experimental results demonstrate its advantages over several other band selection methods.

show abstract

Supervised band selection for optimal use of data from airborne hyperspectral sensors

Cited by 23 publications

References 4 publications

Sparse linear filters for detection and classification in hyperspectral imagery

Sparse linear filters for detection and classification in hyperspectral imagery

Automatic Band Selection Using Spatial-Structure Information and Classifier-Based Clustering

Semisupervised Hyperspectral Band Selection Via Spectral–Spatial Hypergraph Model

Contact Info

Product

Resources

About