An SVDD-based algorithm for target detection in hyperspectral imagery

Wesam Sakla, Andrew Chan, Jim Ji, Adel Sakla

Research output: Contribution to journalArticle

55 Citations (Scopus)

Abstract

Spectral variability remains a challenging problem for target detection and classification in hyperspectral (HS) imagery. In this letter, we have applied the nonlinear support vector data description (SVDD) to perform full-pixel target detection. Using a pure target signature and a first-order Markov model, we have developed a novel pattern recognition algorithm to train an SVDD to characterize the target class. We have inserted target signatures into an urban HS scene with varying levels of spectral variability to explore the performance of the proposed SVDD target detector in different scenarios. The proposed approach makes no assumptions regarding the underlying distribution of the scene data as do traditional stochastic detectors such as the matched filter (MF). Detection results in the form of confusion matrices, and receiver-operating- characteristic curves demonstrate that the proposed SVDD-based algorithm is highly accurate and yields higher true positive rates and lower false positive rates than the MF.

Original languageEnglish
Article number5613915
Pages (from-to)384-388
Number of pages5
JournalIEEE Geoscience and Remote Sensing Letters
Volume8
Issue number2
DOIs
Publication statusPublished - Mar 2011
Externally publishedYes

Fingerprint

Data description
Target tracking
imagery
Matched filters
Detectors
filter
Pattern recognition
pattern recognition
Pixels
train
pixel
detection
matrix

Keywords

  • Automatic target recognition (ATR)
  • hyperspectral imagery
  • support vector data description (SVDD)
  • target detection

ASJC Scopus subject areas

  • Geotechnical Engineering and Engineering Geology
  • Electrical and Electronic Engineering

Cite this

An SVDD-based algorithm for target detection in hyperspectral imagery. / Sakla, Wesam; Chan, Andrew; Ji, Jim; Sakla, Adel.

In: IEEE Geoscience and Remote Sensing Letters, Vol. 8, No. 2, 5613915, 03.2011, p. 384-388.

Research output: Contribution to journalArticle

Sakla, Wesam ; Chan, Andrew ; Ji, Jim ; Sakla, Adel. / An SVDD-based algorithm for target detection in hyperspectral imagery. In: IEEE Geoscience and Remote Sensing Letters. 2011 ; Vol. 8, No. 2. pp. 384-388.
@article{bdbdca438ccd418eb04d4ca4b62bb885,
title = "An SVDD-based algorithm for target detection in hyperspectral imagery",
abstract = "Spectral variability remains a challenging problem for target detection and classification in hyperspectral (HS) imagery. In this letter, we have applied the nonlinear support vector data description (SVDD) to perform full-pixel target detection. Using a pure target signature and a first-order Markov model, we have developed a novel pattern recognition algorithm to train an SVDD to characterize the target class. We have inserted target signatures into an urban HS scene with varying levels of spectral variability to explore the performance of the proposed SVDD target detector in different scenarios. The proposed approach makes no assumptions regarding the underlying distribution of the scene data as do traditional stochastic detectors such as the matched filter (MF). Detection results in the form of confusion matrices, and receiver-operating- characteristic curves demonstrate that the proposed SVDD-based algorithm is highly accurate and yields higher true positive rates and lower false positive rates than the MF.",
keywords = "Automatic target recognition (ATR), hyperspectral imagery, support vector data description (SVDD), target detection",
author = "Wesam Sakla and Andrew Chan and Jim Ji and Adel Sakla",
year = "2011",
month = "3",
doi = "10.1109/LGRS.2010.2078795",
language = "English",
volume = "8",
pages = "384--388",
journal = "IEEE Geoscience and Remote Sensing Letters",
issn = "1545-598X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

TY - JOUR

T1 - An SVDD-based algorithm for target detection in hyperspectral imagery

AU - Sakla, Wesam

AU - Chan, Andrew

AU - Ji, Jim

AU - Sakla, Adel

PY - 2011/3

Y1 - 2011/3

N2 - Spectral variability remains a challenging problem for target detection and classification in hyperspectral (HS) imagery. In this letter, we have applied the nonlinear support vector data description (SVDD) to perform full-pixel target detection. Using a pure target signature and a first-order Markov model, we have developed a novel pattern recognition algorithm to train an SVDD to characterize the target class. We have inserted target signatures into an urban HS scene with varying levels of spectral variability to explore the performance of the proposed SVDD target detector in different scenarios. The proposed approach makes no assumptions regarding the underlying distribution of the scene data as do traditional stochastic detectors such as the matched filter (MF). Detection results in the form of confusion matrices, and receiver-operating- characteristic curves demonstrate that the proposed SVDD-based algorithm is highly accurate and yields higher true positive rates and lower false positive rates than the MF.

AB - Spectral variability remains a challenging problem for target detection and classification in hyperspectral (HS) imagery. In this letter, we have applied the nonlinear support vector data description (SVDD) to perform full-pixel target detection. Using a pure target signature and a first-order Markov model, we have developed a novel pattern recognition algorithm to train an SVDD to characterize the target class. We have inserted target signatures into an urban HS scene with varying levels of spectral variability to explore the performance of the proposed SVDD target detector in different scenarios. The proposed approach makes no assumptions regarding the underlying distribution of the scene data as do traditional stochastic detectors such as the matched filter (MF). Detection results in the form of confusion matrices, and receiver-operating- characteristic curves demonstrate that the proposed SVDD-based algorithm is highly accurate and yields higher true positive rates and lower false positive rates than the MF.

KW - Automatic target recognition (ATR)

KW - hyperspectral imagery

KW - support vector data description (SVDD)

KW - target detection

UR - http://www.scopus.com/inward/record.url?scp=79952039869&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952039869&partnerID=8YFLogxK

U2 - 10.1109/LGRS.2010.2078795

DO - 10.1109/LGRS.2010.2078795

M3 - Article

VL - 8

SP - 384

EP - 388

JO - IEEE Geoscience and Remote Sensing Letters

JF - IEEE Geoscience and Remote Sensing Letters

SN - 1545-598X

IS - 2

M1 - 5613915

ER -