Segmentation of hyperspectral images based on histograms of principal components

J. Silverman; S. R. Rotman; C. E. Caefer

doi:10.1117/12.451537

Segmentation of hyperspectral images based on histograms of principal components

J. Silverman, S. R. Rotman, C. E. Caefer

Department of Electrical & Computer Engineering

Research output: Contribution to journal › Conference article › peer-review

21 Scopus citations

Abstract

Further refinements are presented on a simple and fast way to cluster/segment hyperspectral imagery. In earlier work, it was shown that, starting with the first 2 principal component images, one could form a 2-dimensional histogram and cluster all pixels on the basis of the proximity to the peaks. Issues that we analyzed this year are the proper weighting of the different principal components as a function of the peak shape and automatic methods based on an entropy measure to control the number of clusters and the segmentation of the data to produce the most meaningful results. Examples from both visible and infrared hyperspectral data will be shown.

Original language	English
Pages (from-to)	270-277
Number of pages	8
Journal	Proceedings of SPIE - The International Society for Optical Engineering
Volume	4816
DOIs	https://doi.org/10.1117/12.451537
State	Published - 1 Dec 2002
Event	Imaging Spectrometry VII - Seattle, WA, United States Duration: 8 Jul 2002 → 10 Jul 2002

Keywords

Entropy
Histograms
Hyperspectral
Principal components
Segmentation

ASJC Scopus subject areas

Electronic, Optical and Magnetic Materials
Condensed Matter Physics
Computer Science Applications
Applied Mathematics
Electrical and Electronic Engineering

Access to Document

10.1117/12.451537

Cite this

@article{6692636939ea4c07ae23d6d58fbaf438,

title = "Segmentation of hyperspectral images based on histograms of principal components",

abstract = "Further refinements are presented on a simple and fast way to cluster/segment hyperspectral imagery. In earlier work, it was shown that, starting with the first 2 principal component images, one could form a 2-dimensional histogram and cluster all pixels on the basis of the proximity to the peaks. Issues that we analyzed this year are the proper weighting of the different principal components as a function of the peak shape and automatic methods based on an entropy measure to control the number of clusters and the segmentation of the data to produce the most meaningful results. Examples from both visible and infrared hyperspectral data will be shown.",

keywords = "Entropy, Histograms, Hyperspectral, Principal components, Segmentation",

author = "J. Silverman and Rotman, {S. R.} and Caefer, {C. E.}",

year = "2002",

month = dec,

day = "1",

doi = "10.1117/12.451537",

language = "English",

volume = "4816",

pages = "270--277",

journal = "Proceedings of SPIE - The International Society for Optical Engineering",

issn = "0277-786X",

publisher = "SPIE",

note = "Imaging Spectrometry VII ; Conference date: 08-07-2002 Through 10-07-2002",

}

TY - JOUR

T1 - Segmentation of hyperspectral images based on histograms of principal components

AU - Silverman, J.

AU - Rotman, S. R.

AU - Caefer, C. E.

PY - 2002/12/1

Y1 - 2002/12/1

N2 - Further refinements are presented on a simple and fast way to cluster/segment hyperspectral imagery. In earlier work, it was shown that, starting with the first 2 principal component images, one could form a 2-dimensional histogram and cluster all pixels on the basis of the proximity to the peaks. Issues that we analyzed this year are the proper weighting of the different principal components as a function of the peak shape and automatic methods based on an entropy measure to control the number of clusters and the segmentation of the data to produce the most meaningful results. Examples from both visible and infrared hyperspectral data will be shown.

AB - Further refinements are presented on a simple and fast way to cluster/segment hyperspectral imagery. In earlier work, it was shown that, starting with the first 2 principal component images, one could form a 2-dimensional histogram and cluster all pixels on the basis of the proximity to the peaks. Issues that we analyzed this year are the proper weighting of the different principal components as a function of the peak shape and automatic methods based on an entropy measure to control the number of clusters and the segmentation of the data to produce the most meaningful results. Examples from both visible and infrared hyperspectral data will be shown.

KW - Entropy

KW - Histograms

KW - Hyperspectral

KW - Principal components

KW - Segmentation

UR - http://www.scopus.com/inward/record.url?scp=0036980811&partnerID=8YFLogxK

U2 - 10.1117/12.451537

DO - 10.1117/12.451537

M3 - Conference article

AN - SCOPUS:0036980811

SN - 0277-786X

VL - 4816

SP - 270

EP - 277

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

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

T2 - Imaging Spectrometry VII

Y2 - 8 July 2002 through 10 July 2002

ER -

Segmentation of hyperspectral images based on histograms of principal components

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this