New developments in target detection in hyperspectral imagery using spectral metrics and spectra extraction

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

2 Citations (Scopus)

Abstract

An important problem in processing multispectral / hyperspectral imagery consists in the design of methods for the detection of targets. In this paper we investigate a class of detection filters based on band and spectral selection. Band selection (band screening) refers to searching for the spectral bands that would yield the largest separation between target and background. Spectral selection (spectral screening) is a technique used in reducing the multispectral / hyperspectral data to a representative subset of spectra. The subset is formed such that any two spectra in it are dissimilar and, for any spectrum in the original image cube, there is a similar spectrum in the subset. Spectral screening is performed in a sequential manner, at each step, the subset being increased with a spectrum dissimilar from all the spectra already selected. We modified the spectral screening algorithm such that at the selection step the spectrum with the largest distance from the set is selected. While not introducing additional computational complexity, the Maximum Spectral Screening (MSS) algorithm ensures that the overlap among the representatives is minimized. The detection filters were obtained as the classification projector matrices based on the spectral subset. The developed algorithms were tested on HYDICE hyperspectral data using the spectral angle and the spectral information divergence. The results indicate that regular MSS outperforms band selection followed by MSS, both classes outperforming regular spectral screening.

Original languageEnglish
Title of host publicationAmerican Society for Photogrammetry and Remote Sensing - ASPRS Annual Conference 2007
Subtitle of host publicationIdentifying Geospatial Solutions
Pages604-614
Number of pages11
Volume2
StatePublished - 1 Dec 2007
EventASPRS Annual Conference 2007: Identifying Geospatial Solutions - Tampa, FL, United States
Duration: 7 May 200711 May 2007

Other

OtherASPRS Annual Conference 2007: Identifying Geospatial Solutions
CountryUnited States
CityTampa, FL
Period7/05/0711/05/07

Fingerprint

Target tracking
Screening
imagery
filter
screening
detection
Set theory
Computational complexity
divergence
matrix
Processing

Cite this

Robila, S. (2007). New developments in target detection in hyperspectral imagery using spectral metrics and spectra extraction. In American Society for Photogrammetry and Remote Sensing - ASPRS Annual Conference 2007: Identifying Geospatial Solutions (Vol. 2, pp. 604-614)
Robila, Stefan. / New developments in target detection in hyperspectral imagery using spectral metrics and spectra extraction. American Society for Photogrammetry and Remote Sensing - ASPRS Annual Conference 2007: Identifying Geospatial Solutions. Vol. 2 2007. pp. 604-614
@inproceedings{37af3dccff8942f99b0413aeb05ada5f,
title = "New developments in target detection in hyperspectral imagery using spectral metrics and spectra extraction",
abstract = "An important problem in processing multispectral / hyperspectral imagery consists in the design of methods for the detection of targets. In this paper we investigate a class of detection filters based on band and spectral selection. Band selection (band screening) refers to searching for the spectral bands that would yield the largest separation between target and background. Spectral selection (spectral screening) is a technique used in reducing the multispectral / hyperspectral data to a representative subset of spectra. The subset is formed such that any two spectra in it are dissimilar and, for any spectrum in the original image cube, there is a similar spectrum in the subset. Spectral screening is performed in a sequential manner, at each step, the subset being increased with a spectrum dissimilar from all the spectra already selected. We modified the spectral screening algorithm such that at the selection step the spectrum with the largest distance from the set is selected. While not introducing additional computational complexity, the Maximum Spectral Screening (MSS) algorithm ensures that the overlap among the representatives is minimized. The detection filters were obtained as the classification projector matrices based on the spectral subset. The developed algorithms were tested on HYDICE hyperspectral data using the spectral angle and the spectral information divergence. The results indicate that regular MSS outperforms band selection followed by MSS, both classes outperforming regular spectral screening.",
author = "Stefan Robila",
year = "2007",
month = "12",
day = "1",
language = "English",
isbn = "9781604232240",
volume = "2",
pages = "604--614",
booktitle = "American Society for Photogrammetry and Remote Sensing - ASPRS Annual Conference 2007",

}

Robila, S 2007, New developments in target detection in hyperspectral imagery using spectral metrics and spectra extraction. in American Society for Photogrammetry and Remote Sensing - ASPRS Annual Conference 2007: Identifying Geospatial Solutions. vol. 2, pp. 604-614, ASPRS Annual Conference 2007: Identifying Geospatial Solutions, Tampa, FL, United States, 7/05/07.

New developments in target detection in hyperspectral imagery using spectral metrics and spectra extraction. / Robila, Stefan.

American Society for Photogrammetry and Remote Sensing - ASPRS Annual Conference 2007: Identifying Geospatial Solutions. Vol. 2 2007. p. 604-614.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - New developments in target detection in hyperspectral imagery using spectral metrics and spectra extraction

AU - Robila, Stefan

PY - 2007/12/1

Y1 - 2007/12/1

N2 - An important problem in processing multispectral / hyperspectral imagery consists in the design of methods for the detection of targets. In this paper we investigate a class of detection filters based on band and spectral selection. Band selection (band screening) refers to searching for the spectral bands that would yield the largest separation between target and background. Spectral selection (spectral screening) is a technique used in reducing the multispectral / hyperspectral data to a representative subset of spectra. The subset is formed such that any two spectra in it are dissimilar and, for any spectrum in the original image cube, there is a similar spectrum in the subset. Spectral screening is performed in a sequential manner, at each step, the subset being increased with a spectrum dissimilar from all the spectra already selected. We modified the spectral screening algorithm such that at the selection step the spectrum with the largest distance from the set is selected. While not introducing additional computational complexity, the Maximum Spectral Screening (MSS) algorithm ensures that the overlap among the representatives is minimized. The detection filters were obtained as the classification projector matrices based on the spectral subset. The developed algorithms were tested on HYDICE hyperspectral data using the spectral angle and the spectral information divergence. The results indicate that regular MSS outperforms band selection followed by MSS, both classes outperforming regular spectral screening.

AB - An important problem in processing multispectral / hyperspectral imagery consists in the design of methods for the detection of targets. In this paper we investigate a class of detection filters based on band and spectral selection. Band selection (band screening) refers to searching for the spectral bands that would yield the largest separation between target and background. Spectral selection (spectral screening) is a technique used in reducing the multispectral / hyperspectral data to a representative subset of spectra. The subset is formed such that any two spectra in it are dissimilar and, for any spectrum in the original image cube, there is a similar spectrum in the subset. Spectral screening is performed in a sequential manner, at each step, the subset being increased with a spectrum dissimilar from all the spectra already selected. We modified the spectral screening algorithm such that at the selection step the spectrum with the largest distance from the set is selected. While not introducing additional computational complexity, the Maximum Spectral Screening (MSS) algorithm ensures that the overlap among the representatives is minimized. The detection filters were obtained as the classification projector matrices based on the spectral subset. The developed algorithms were tested on HYDICE hyperspectral data using the spectral angle and the spectral information divergence. The results indicate that regular MSS outperforms band selection followed by MSS, both classes outperforming regular spectral screening.

UR - http://www.scopus.com/inward/record.url?scp=84868622236&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781604232240

VL - 2

SP - 604

EP - 614

BT - American Society for Photogrammetry and Remote Sensing - ASPRS Annual Conference 2007

ER -

Robila S. New developments in target detection in hyperspectral imagery using spectral metrics and spectra extraction. In American Society for Photogrammetry and Remote Sensing - ASPRS Annual Conference 2007: Identifying Geospatial Solutions. Vol. 2. 2007. p. 604-614