skip to main content

DOE PAGESDOE PAGES

Title: Novel Histogram Based Unsupervised Classification Technique to Determine Natural Classes From Biophysically Relevant Fit Parameters to Hyperspectral Data

Hyperspectral image analysis has benefited from an array of methods that take advantage of the increased spectral depth compared to multispectral sensors; however, the focus of these developments has been on supervised classification methods. Lack of a priori knowledge regarding land cover characteristics can make unsupervised classification methods preferable under certain circumstances. An unsupervised classification technique is presented in this paper that utilizes physically relevant basis functions to model the reflectance spectra. These fit parameters used to generate the basis functions allow clustering based on spectral characteristics rather than spectral channels and provide both noise and data reduction. Histogram splitting of the fit parameters is then used as a means of producing an unsupervised classification. Unlike current unsupervised classification techniques that rely primarily on Euclidian distance measures to determine similarity, the unsupervised classification technique uses the natural splitting of the fit parameters associated with the basis functions creating clusters that are similar in terms of physical parameters. The data set used in this work utilizes the publicly available data collected at Indian Pines, Indiana. This data set provides reference data allowing for comparisons of the efficacy of different unsupervised data analysis. The unsupervised histogram splitting technique presented in this papermore » is shown to be better than the standard unsupervised ISODATA clustering technique with an overall accuracy of 34.3/19.0% before merging and 40.9/39.2% after merging. Finally, this improvement is also seen as an improvement of kappa before/after merging of 24.8/30.5 for the histogram splitting technique compared to 15.8/28.5 for ISODATA.« less
Authors:
 [1] ;  [1] ;  [1] ;  [1] ;  [1]
  1. Montana State Univ., Bozeman, MT (United States)
Publication Date:
Grant/Contract Number:
FC26-05NT42587
Type:
Accepted Manuscript
Journal Name:
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Additional Journal Information:
Journal Volume: 10; Journal Issue: 9; Journal ID: ISSN 1939-1404
Publisher:
IEEE
Research Org:
Montana State Univ., Bozeman, MT (United States)
Sponsoring Org:
USDOE Office of Fossil Energy (FE)
Country of Publication:
United States
Language:
English
Subject:
58 GEOSCIENCES; histograms; vegetation mapping; training; hyperspectral imaging; data models; sensors; agriculture; biophysics; clustering methods; remote sensing
OSTI Identifier:
1409671

McCann, Cooper, Repasky, Kevin S., Morin, Mikindra, Lawrence, Rick L., and Powell, Scott. Novel Histogram Based Unsupervised Classification Technique to Determine Natural Classes From Biophysically Relevant Fit Parameters to Hyperspectral Data. United States: N. p., Web. doi:10.1109/JSTARS.2017.2701360.
McCann, Cooper, Repasky, Kevin S., Morin, Mikindra, Lawrence, Rick L., & Powell, Scott. Novel Histogram Based Unsupervised Classification Technique to Determine Natural Classes From Biophysically Relevant Fit Parameters to Hyperspectral Data. United States. doi:10.1109/JSTARS.2017.2701360.
McCann, Cooper, Repasky, Kevin S., Morin, Mikindra, Lawrence, Rick L., and Powell, Scott. 2017. "Novel Histogram Based Unsupervised Classification Technique to Determine Natural Classes From Biophysically Relevant Fit Parameters to Hyperspectral Data". United States. doi:10.1109/JSTARS.2017.2701360. https://www.osti.gov/servlets/purl/1409671.
@article{osti_1409671,
title = {Novel Histogram Based Unsupervised Classification Technique to Determine Natural Classes From Biophysically Relevant Fit Parameters to Hyperspectral Data},
author = {McCann, Cooper and Repasky, Kevin S. and Morin, Mikindra and Lawrence, Rick L. and Powell, Scott},
abstractNote = {Hyperspectral image analysis has benefited from an array of methods that take advantage of the increased spectral depth compared to multispectral sensors; however, the focus of these developments has been on supervised classification methods. Lack of a priori knowledge regarding land cover characteristics can make unsupervised classification methods preferable under certain circumstances. An unsupervised classification technique is presented in this paper that utilizes physically relevant basis functions to model the reflectance spectra. These fit parameters used to generate the basis functions allow clustering based on spectral characteristics rather than spectral channels and provide both noise and data reduction. Histogram splitting of the fit parameters is then used as a means of producing an unsupervised classification. Unlike current unsupervised classification techniques that rely primarily on Euclidian distance measures to determine similarity, the unsupervised classification technique uses the natural splitting of the fit parameters associated with the basis functions creating clusters that are similar in terms of physical parameters. The data set used in this work utilizes the publicly available data collected at Indian Pines, Indiana. This data set provides reference data allowing for comparisons of the efficacy of different unsupervised data analysis. The unsupervised histogram splitting technique presented in this paper is shown to be better than the standard unsupervised ISODATA clustering technique with an overall accuracy of 34.3/19.0% before merging and 40.9/39.2% after merging. Finally, this improvement is also seen as an improvement of kappa before/after merging of 24.8/30.5 for the histogram splitting technique compared to 15.8/28.5 for ISODATA.},
doi = {10.1109/JSTARS.2017.2701360},
journal = {IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},
number = 9,
volume = 10,
place = {United States},
year = {2017},
month = {5}
}