Real time processing of hyperspectral images

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

1 Citation (Scopus)

Abstract

We describe the development of a real-time processing tool for hyperspectral imagery based on off-the-shelf equipment and higher level programming language implementation (C++ and Java). The algorithms we developed are derived from previously introduced spectra matching and feature extraction tools. The first group is based on spectra identification and spectral screening, a method that allows the identification of representative spectra from a data set. The second group is based on Principal Component Analysis (PCA) and Independent Component Analysis (ICA). When applied to multidimensional data, PCA linearly transforms them such that the resulting components are uncorrelated and their variance maximized. In ICA, given a linear mixture of statistical independent sources, the goal is to recover these components by producing an unmixing matrix. The effectiveness of the proposed real time algorithms were tested on an in-house system composed of a commercially available hyperspectral camera and a multiprocessor computer system. Preliminary results targeted at the feasibility of the tool show that reasonable accuracy can be maintained in the real time requirements. The described project supports the further development of hyperspectral imaging as a general tool in remote sensing.

Original languageEnglish
Title of host publicationAmerican Society for Photogrammetry and Remote Sensing - Annual Conference of the American Society for Photogrammetry and Remote Sensing 2006
Subtitle of host publicationProspecting for Geospatial Information Integration
Pages1196-1203
Number of pages8
Volume3
StatePublished - 1 Dec 2006
EventAnnual Conference of the American Society for Photogrammetry and Remote Sensing 2006: Prospecting for Geospatial Information Integration, ASPRS 2006 - Reno, NV, United States
Duration: 1 May 20065 May 2006

Other

OtherAnnual Conference of the American Society for Photogrammetry and Remote Sensing 2006: Prospecting for Geospatial Information Integration, ASPRS 2006
CountryUnited States
CityReno, NV
Period1/05/065/05/06

Fingerprint

principal component analysis
Independent component analysis
Processing
Principal component analysis
computer system
Java programming language
imagery
transform
remote sensing
Feature extraction
Remote sensing
Screening
matrix
Computer systems
Cameras
analysis
project
method
screening
Hyperspectral imaging

Cite this

Robila, S. (2006). Real time processing of hyperspectral images. In American Society for Photogrammetry and Remote Sensing - Annual Conference of the American Society for Photogrammetry and Remote Sensing 2006: Prospecting for Geospatial Information Integration (Vol. 3, pp. 1196-1203)
Robila, Stefan. / Real time processing of hyperspectral images. American Society for Photogrammetry and Remote Sensing - Annual Conference of the American Society for Photogrammetry and Remote Sensing 2006: Prospecting for Geospatial Information Integration. Vol. 3 2006. pp. 1196-1203
@inproceedings{1ecb6f72197047d3bc2e57179984a196,
title = "Real time processing of hyperspectral images",
abstract = "We describe the development of a real-time processing tool for hyperspectral imagery based on off-the-shelf equipment and higher level programming language implementation (C++ and Java). The algorithms we developed are derived from previously introduced spectra matching and feature extraction tools. The first group is based on spectra identification and spectral screening, a method that allows the identification of representative spectra from a data set. The second group is based on Principal Component Analysis (PCA) and Independent Component Analysis (ICA). When applied to multidimensional data, PCA linearly transforms them such that the resulting components are uncorrelated and their variance maximized. In ICA, given a linear mixture of statistical independent sources, the goal is to recover these components by producing an unmixing matrix. The effectiveness of the proposed real time algorithms were tested on an in-house system composed of a commercially available hyperspectral camera and a multiprocessor computer system. Preliminary results targeted at the feasibility of the tool show that reasonable accuracy can be maintained in the real time requirements. The described project supports the further development of hyperspectral imaging as a general tool in remote sensing.",
author = "Stefan Robila",
year = "2006",
month = "12",
day = "1",
language = "English",
isbn = "9781604237290",
volume = "3",
pages = "1196--1203",
booktitle = "American Society for Photogrammetry and Remote Sensing - Annual Conference of the American Society for Photogrammetry and Remote Sensing 2006",

}

Robila, S 2006, Real time processing of hyperspectral images. in American Society for Photogrammetry and Remote Sensing - Annual Conference of the American Society for Photogrammetry and Remote Sensing 2006: Prospecting for Geospatial Information Integration. vol. 3, pp. 1196-1203, Annual Conference of the American Society for Photogrammetry and Remote Sensing 2006: Prospecting for Geospatial Information Integration, ASPRS 2006, Reno, NV, United States, 1/05/06.

Real time processing of hyperspectral images. / Robila, Stefan.

American Society for Photogrammetry and Remote Sensing - Annual Conference of the American Society for Photogrammetry and Remote Sensing 2006: Prospecting for Geospatial Information Integration. Vol. 3 2006. p. 1196-1203.

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

TY - GEN

T1 - Real time processing of hyperspectral images

AU - Robila, Stefan

PY - 2006/12/1

Y1 - 2006/12/1

N2 - We describe the development of a real-time processing tool for hyperspectral imagery based on off-the-shelf equipment and higher level programming language implementation (C++ and Java). The algorithms we developed are derived from previously introduced spectra matching and feature extraction tools. The first group is based on spectra identification and spectral screening, a method that allows the identification of representative spectra from a data set. The second group is based on Principal Component Analysis (PCA) and Independent Component Analysis (ICA). When applied to multidimensional data, PCA linearly transforms them such that the resulting components are uncorrelated and their variance maximized. In ICA, given a linear mixture of statistical independent sources, the goal is to recover these components by producing an unmixing matrix. The effectiveness of the proposed real time algorithms were tested on an in-house system composed of a commercially available hyperspectral camera and a multiprocessor computer system. Preliminary results targeted at the feasibility of the tool show that reasonable accuracy can be maintained in the real time requirements. The described project supports the further development of hyperspectral imaging as a general tool in remote sensing.

AB - We describe the development of a real-time processing tool for hyperspectral imagery based on off-the-shelf equipment and higher level programming language implementation (C++ and Java). The algorithms we developed are derived from previously introduced spectra matching and feature extraction tools. The first group is based on spectra identification and spectral screening, a method that allows the identification of representative spectra from a data set. The second group is based on Principal Component Analysis (PCA) and Independent Component Analysis (ICA). When applied to multidimensional data, PCA linearly transforms them such that the resulting components are uncorrelated and their variance maximized. In ICA, given a linear mixture of statistical independent sources, the goal is to recover these components by producing an unmixing matrix. The effectiveness of the proposed real time algorithms were tested on an in-house system composed of a commercially available hyperspectral camera and a multiprocessor computer system. Preliminary results targeted at the feasibility of the tool show that reasonable accuracy can be maintained in the real time requirements. The described project supports the further development of hyperspectral imaging as a general tool in remote sensing.

UR - http://www.scopus.com/inward/record.url?scp=84869019738&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781604237290

VL - 3

SP - 1196

EP - 1203

BT - American Society for Photogrammetry and Remote Sensing - Annual Conference of the American Society for Photogrammetry and Remote Sensing 2006

ER -

Robila S. Real time processing of hyperspectral images. In American Society for Photogrammetry and Remote Sensing - Annual Conference of the American Society for Photogrammetry and Remote Sensing 2006: Prospecting for Geospatial Information Integration. Vol. 3. 2006. p. 1196-1203