Methodology for hyperspectral band selection

Peter Bajcsy, Peter Groves

Research output: Contribution to journalReview article

Abstract

While hyperspectral data are very rich in information, processing the hyperspectral data poses several challenges regarding computational requirements, information redundancy removal, relevant information identification, and modeling accuracy. In this paper we present a new methodology for combining unsupervised and supervised methods under classification accuracy and computational requirement constraints that is designed to perform hyperspectral band (wavelength range) selection and statistical modeling method selection. The band and method selections are utilized for prediction of continuous ground variables using airborne hyperspectral measurements. The novelty of the proposed work is in combining strengths of unsupervised and supervised band selection methods to build a computationally efficient and accurate band selection system. The unsupervised methods are used to rank hyperspectral bands while the accuracy of the predictions of supervised methods are used to score those rankings. We conducted experiments with seven unsupervised and three supervised methods. The list of unsupervised methods includes information entropy, first and second spectral derivative, spatial contrast, spectral ratio, correlation, and principal component analysis ranking combined with regression, regression tree, and instance-based supervised methods. These methods were applied to a data set that relates ground measurements of soil electrical conductivity with airborne hyperspectral image values. The outcomes of our analysis led to a conclusion that the optimum number of bands in this domain is the top four to eight bands obtained by the entropy unsupervised method followed by the regression tree supervised method evaluation. Although the proposed band selection approach is demonstrated with a data set from the precision agriculture domain, it applies in other hyperspectral application domains.

Original languageEnglish (US)
Pages (from-to)793-802
Number of pages10
JournalPhotogrammetric Engineering and Remote Sensing
Volume70
Issue number7
DOIs
StatePublished - Jul 1 2004

Fingerprint

Entropy
methodology
Principal component analysis
Agriculture
Redundancy
Derivatives
Soils
Wavelength
Experiments
ranking
entropy
method
precision agriculture
information processing
prediction
modeling
electrical conductivity
principal component analysis
Electric Conductivity
wavelength

ASJC Scopus subject areas

  • Computers in Earth Sciences

Cite this

Methodology for hyperspectral band selection. / Bajcsy, Peter; Groves, Peter.

In: Photogrammetric Engineering and Remote Sensing, Vol. 70, No. 7, 01.07.2004, p. 793-802.

Research output: Contribution to journalReview article

@article{4ca855349b86421a88571fdfe5cceb02,
title = "Methodology for hyperspectral band selection",
abstract = "While hyperspectral data are very rich in information, processing the hyperspectral data poses several challenges regarding computational requirements, information redundancy removal, relevant information identification, and modeling accuracy. In this paper we present a new methodology for combining unsupervised and supervised methods under classification accuracy and computational requirement constraints that is designed to perform hyperspectral band (wavelength range) selection and statistical modeling method selection. The band and method selections are utilized for prediction of continuous ground variables using airborne hyperspectral measurements. The novelty of the proposed work is in combining strengths of unsupervised and supervised band selection methods to build a computationally efficient and accurate band selection system. The unsupervised methods are used to rank hyperspectral bands while the accuracy of the predictions of supervised methods are used to score those rankings. We conducted experiments with seven unsupervised and three supervised methods. The list of unsupervised methods includes information entropy, first and second spectral derivative, spatial contrast, spectral ratio, correlation, and principal component analysis ranking combined with regression, regression tree, and instance-based supervised methods. These methods were applied to a data set that relates ground measurements of soil electrical conductivity with airborne hyperspectral image values. The outcomes of our analysis led to a conclusion that the optimum number of bands in this domain is the top four to eight bands obtained by the entropy unsupervised method followed by the regression tree supervised method evaluation. Although the proposed band selection approach is demonstrated with a data set from the precision agriculture domain, it applies in other hyperspectral application domains.",
author = "Peter Bajcsy and Peter Groves",
year = "2004",
month = "7",
day = "1",
doi = "10.14358/PERS.70.7.793",
language = "English (US)",
volume = "70",
pages = "793--802",
journal = "Photogrammetric Engineering and Remote Sensing",
issn = "0099-1112",
publisher = "American Society for Photogrammetry and Remote Sensing",
number = "7",

}

TY - JOUR

T1 - Methodology for hyperspectral band selection

AU - Bajcsy, Peter

AU - Groves, Peter

PY - 2004/7/1

Y1 - 2004/7/1

N2 - While hyperspectral data are very rich in information, processing the hyperspectral data poses several challenges regarding computational requirements, information redundancy removal, relevant information identification, and modeling accuracy. In this paper we present a new methodology for combining unsupervised and supervised methods under classification accuracy and computational requirement constraints that is designed to perform hyperspectral band (wavelength range) selection and statistical modeling method selection. The band and method selections are utilized for prediction of continuous ground variables using airborne hyperspectral measurements. The novelty of the proposed work is in combining strengths of unsupervised and supervised band selection methods to build a computationally efficient and accurate band selection system. The unsupervised methods are used to rank hyperspectral bands while the accuracy of the predictions of supervised methods are used to score those rankings. We conducted experiments with seven unsupervised and three supervised methods. The list of unsupervised methods includes information entropy, first and second spectral derivative, spatial contrast, spectral ratio, correlation, and principal component analysis ranking combined with regression, regression tree, and instance-based supervised methods. These methods were applied to a data set that relates ground measurements of soil electrical conductivity with airborne hyperspectral image values. The outcomes of our analysis led to a conclusion that the optimum number of bands in this domain is the top four to eight bands obtained by the entropy unsupervised method followed by the regression tree supervised method evaluation. Although the proposed band selection approach is demonstrated with a data set from the precision agriculture domain, it applies in other hyperspectral application domains.

AB - While hyperspectral data are very rich in information, processing the hyperspectral data poses several challenges regarding computational requirements, information redundancy removal, relevant information identification, and modeling accuracy. In this paper we present a new methodology for combining unsupervised and supervised methods under classification accuracy and computational requirement constraints that is designed to perform hyperspectral band (wavelength range) selection and statistical modeling method selection. The band and method selections are utilized for prediction of continuous ground variables using airborne hyperspectral measurements. The novelty of the proposed work is in combining strengths of unsupervised and supervised band selection methods to build a computationally efficient and accurate band selection system. The unsupervised methods are used to rank hyperspectral bands while the accuracy of the predictions of supervised methods are used to score those rankings. We conducted experiments with seven unsupervised and three supervised methods. The list of unsupervised methods includes information entropy, first and second spectral derivative, spatial contrast, spectral ratio, correlation, and principal component analysis ranking combined with regression, regression tree, and instance-based supervised methods. These methods were applied to a data set that relates ground measurements of soil electrical conductivity with airborne hyperspectral image values. The outcomes of our analysis led to a conclusion that the optimum number of bands in this domain is the top four to eight bands obtained by the entropy unsupervised method followed by the regression tree supervised method evaluation. Although the proposed band selection approach is demonstrated with a data set from the precision agriculture domain, it applies in other hyperspectral application domains.

UR - http://www.scopus.com/inward/record.url?scp=4143064738&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=4143064738&partnerID=8YFLogxK

U2 - 10.14358/PERS.70.7.793

DO - 10.14358/PERS.70.7.793

M3 - Review article

AN - SCOPUS:4143064738

VL - 70

SP - 793

EP - 802

JO - Photogrammetric Engineering and Remote Sensing

JF - Photogrammetric Engineering and Remote Sensing

SN - 0099-1112

IS - 7

ER -