Segmentation-based detection of targets in foliage penetrating SAR images

Amit Banerjee, Philippe Burlina

Research output: Contribution to journalConference article

Abstract

Segmentation and labeling algorithms for foliage penetrating (FOPEN) ultra-wideband Synthetic Aperture Radar (UWB SAR) images are critical components in providing local context in automatic target recognition algorithms. We develop a statistical estimation-theoretic approach to segmenting and labeling the FOPEN images into foliage and non-foliage regions. The labeled maps enable the use of region-adaptive detectors, such as a constant false-alarm rate detector with region-dependent parameters. Segmentation of the images is achieved by performing a maximum a posteriori (MAP) estimate of the pixel labels. By modeling the conditional distribution with a Symmetric Alpha-Stable density and assuming a Markov random field model for the pixel labels, the resulting posterior probability density function is maximized by using simulated annealing to yield the MAP estimate.

Original languageEnglish (US)
Pages (from-to)143-152
Number of pages10
JournalProceedings of SPIE - The International Society for Optical Engineering
Volume3066
DOIs
StatePublished - Dec 1 1997
Externally publishedYes
EventRadar Sensor Technology II - Orlando, FL, United States
Duration: Apr 24 1997Apr 24 1997

Fingerprint

foliage
SAR Image
A Posteriori Estimates
Segmentation
Maximum a Posteriori
Labeling
marking
Target
Labels
Pixel
Pixels
pixels
Detector
Labeling Algorithm
Detectors
Automatic target recognition
Statistical Estimation
False Alarm Rate
Target Recognition
target recognition

Keywords

  • ATR
  • Focus of Attention
  • Foliage penetrating SAR
  • MAP segmentation
  • Region-adaptive CFAR

ASJC Scopus subject areas

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

Cite this

Segmentation-based detection of targets in foliage penetrating SAR images. / Banerjee, Amit; Burlina, Philippe.

In: Proceedings of SPIE - The International Society for Optical Engineering, Vol. 3066, 01.12.1997, p. 143-152.

Research output: Contribution to journalConference article

@article{86483e6835bc40fbb102cdcdcf4dd4b6,
title = "Segmentation-based detection of targets in foliage penetrating SAR images",
abstract = "Segmentation and labeling algorithms for foliage penetrating (FOPEN) ultra-wideband Synthetic Aperture Radar (UWB SAR) images are critical components in providing local context in automatic target recognition algorithms. We develop a statistical estimation-theoretic approach to segmenting and labeling the FOPEN images into foliage and non-foliage regions. The labeled maps enable the use of region-adaptive detectors, such as a constant false-alarm rate detector with region-dependent parameters. Segmentation of the images is achieved by performing a maximum a posteriori (MAP) estimate of the pixel labels. By modeling the conditional distribution with a Symmetric Alpha-Stable density and assuming a Markov random field model for the pixel labels, the resulting posterior probability density function is maximized by using simulated annealing to yield the MAP estimate.",
keywords = "ATR, Focus of Attention, Foliage penetrating SAR, MAP segmentation, Region-adaptive CFAR",
author = "Amit Banerjee and Philippe Burlina",
year = "1997",
month = "12",
day = "1",
doi = "10.1117/12.276097",
language = "English (US)",
volume = "3066",
pages = "143--152",
journal = "Proceedings of SPIE - The International Society for Optical Engineering",
issn = "0277-786X",
publisher = "SPIE",

}

TY - JOUR

T1 - Segmentation-based detection of targets in foliage penetrating SAR images

AU - Banerjee, Amit

AU - Burlina, Philippe

PY - 1997/12/1

Y1 - 1997/12/1

N2 - Segmentation and labeling algorithms for foliage penetrating (FOPEN) ultra-wideband Synthetic Aperture Radar (UWB SAR) images are critical components in providing local context in automatic target recognition algorithms. We develop a statistical estimation-theoretic approach to segmenting and labeling the FOPEN images into foliage and non-foliage regions. The labeled maps enable the use of region-adaptive detectors, such as a constant false-alarm rate detector with region-dependent parameters. Segmentation of the images is achieved by performing a maximum a posteriori (MAP) estimate of the pixel labels. By modeling the conditional distribution with a Symmetric Alpha-Stable density and assuming a Markov random field model for the pixel labels, the resulting posterior probability density function is maximized by using simulated annealing to yield the MAP estimate.

AB - Segmentation and labeling algorithms for foliage penetrating (FOPEN) ultra-wideband Synthetic Aperture Radar (UWB SAR) images are critical components in providing local context in automatic target recognition algorithms. We develop a statistical estimation-theoretic approach to segmenting and labeling the FOPEN images into foliage and non-foliage regions. The labeled maps enable the use of region-adaptive detectors, such as a constant false-alarm rate detector with region-dependent parameters. Segmentation of the images is achieved by performing a maximum a posteriori (MAP) estimate of the pixel labels. By modeling the conditional distribution with a Symmetric Alpha-Stable density and assuming a Markov random field model for the pixel labels, the resulting posterior probability density function is maximized by using simulated annealing to yield the MAP estimate.

KW - ATR

KW - Focus of Attention

KW - Foliage penetrating SAR

KW - MAP segmentation

KW - Region-adaptive CFAR

UR - http://www.scopus.com/inward/record.url?scp=58749116658&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58749116658&partnerID=8YFLogxK

U2 - 10.1117/12.276097

DO - 10.1117/12.276097

M3 - Conference article

AN - SCOPUS:58749116658

VL - 3066

SP - 143

EP - 152

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

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

SN - 0277-786X

ER -