Classification of hyperspectral spatial/spectral patterns using Gauss-Markov random fields

Heidi A. Smartt, J Scott Tyo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Hyperspectral imaging sensors capture digital images in hundreds of contiguous spectral bands, allowing remote material identification. Most algorithms for identifying materials characterize the materials according to spectral information only, ignoring potentially valuable spatial relationships. This paper investigates the use of integrated spatial and spectral information for characterizing materials. It examines the specific situation where a set of pixels has resolution such that it contains spatial patterns of mixed pixels. An autoregressive Gauss-Markov random field (GMRF) is used to model the predictability of a target pixel from neighboring pixels. At the resolution of interest, the GMRF model can successfully classify spatial patterns of aircraft and a residential area from the HYDICE airborne sensor Desert Radiance field collection at Davis Monthan Air Force Base, Arizona.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
Volume6233 I
DOIs
StatePublished - 2006
Externally publishedYes
EventAlgorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XII - Kissimmee, FL, United States
Duration: Apr 17 2006Apr 20 2006

Other

OtherAlgorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XII
CountryUnited States
CityKissimmee, FL
Period4/17/064/20/06

Fingerprint

Pixels
pixels
residential areas
deserts
sensors
Sensors
spectral bands
radiance
aircraft
Aircraft
air
Air

Keywords

  • Gauss-Markov random fields
  • Hyperspectral classification
  • Hyperspectral texture

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Smartt, H. A., & Tyo, J. S. (2006). Classification of hyperspectral spatial/spectral patterns using Gauss-Markov random fields. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 6233 I). [62330I] https://doi.org/10.1117/12.666041

Classification of hyperspectral spatial/spectral patterns using Gauss-Markov random fields. / Smartt, Heidi A.; Tyo, J Scott.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 6233 I 2006. 62330I.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Smartt, HA & Tyo, JS 2006, Classification of hyperspectral spatial/spectral patterns using Gauss-Markov random fields. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 6233 I, 62330I, Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XII, Kissimmee, FL, United States, 4/17/06. https://doi.org/10.1117/12.666041
Smartt HA, Tyo JS. Classification of hyperspectral spatial/spectral patterns using Gauss-Markov random fields. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 6233 I. 2006. 62330I https://doi.org/10.1117/12.666041
Smartt, Heidi A. ; Tyo, J Scott. / Classification of hyperspectral spatial/spectral patterns using Gauss-Markov random fields. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 6233 I 2006.
@inproceedings{300ba76f4fcc4072ad88dadcb1a58bf5,
title = "Classification of hyperspectral spatial/spectral patterns using Gauss-Markov random fields",
abstract = "Hyperspectral imaging sensors capture digital images in hundreds of contiguous spectral bands, allowing remote material identification. Most algorithms for identifying materials characterize the materials according to spectral information only, ignoring potentially valuable spatial relationships. This paper investigates the use of integrated spatial and spectral information for characterizing materials. It examines the specific situation where a set of pixels has resolution such that it contains spatial patterns of mixed pixels. An autoregressive Gauss-Markov random field (GMRF) is used to model the predictability of a target pixel from neighboring pixels. At the resolution of interest, the GMRF model can successfully classify spatial patterns of aircraft and a residential area from the HYDICE airborne sensor Desert Radiance field collection at Davis Monthan Air Force Base, Arizona.",
keywords = "Gauss-Markov random fields, Hyperspectral classification, Hyperspectral texture",
author = "Smartt, {Heidi A.} and Tyo, {J Scott}",
year = "2006",
doi = "10.1117/12.666041",
language = "English (US)",
isbn = "0819462896",
volume = "6233 I",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Classification of hyperspectral spatial/spectral patterns using Gauss-Markov random fields

AU - Smartt, Heidi A.

AU - Tyo, J Scott

PY - 2006

Y1 - 2006

N2 - Hyperspectral imaging sensors capture digital images in hundreds of contiguous spectral bands, allowing remote material identification. Most algorithms for identifying materials characterize the materials according to spectral information only, ignoring potentially valuable spatial relationships. This paper investigates the use of integrated spatial and spectral information for characterizing materials. It examines the specific situation where a set of pixels has resolution such that it contains spatial patterns of mixed pixels. An autoregressive Gauss-Markov random field (GMRF) is used to model the predictability of a target pixel from neighboring pixels. At the resolution of interest, the GMRF model can successfully classify spatial patterns of aircraft and a residential area from the HYDICE airborne sensor Desert Radiance field collection at Davis Monthan Air Force Base, Arizona.

AB - Hyperspectral imaging sensors capture digital images in hundreds of contiguous spectral bands, allowing remote material identification. Most algorithms for identifying materials characterize the materials according to spectral information only, ignoring potentially valuable spatial relationships. This paper investigates the use of integrated spatial and spectral information for characterizing materials. It examines the specific situation where a set of pixels has resolution such that it contains spatial patterns of mixed pixels. An autoregressive Gauss-Markov random field (GMRF) is used to model the predictability of a target pixel from neighboring pixels. At the resolution of interest, the GMRF model can successfully classify spatial patterns of aircraft and a residential area from the HYDICE airborne sensor Desert Radiance field collection at Davis Monthan Air Force Base, Arizona.

KW - Gauss-Markov random fields

KW - Hyperspectral classification

KW - Hyperspectral texture

UR - http://www.scopus.com/inward/record.url?scp=33748635216&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33748635216&partnerID=8YFLogxK

U2 - 10.1117/12.666041

DO - 10.1117/12.666041

M3 - Conference contribution

AN - SCOPUS:33748635216

SN - 0819462896

SN - 9780819462893

VL - 6233 I

BT - Proceedings of SPIE - The International Society for Optical Engineering

ER -