Surface Water Body Detection in Polarimetric SAR Data Using Contextual Complex Wishart Classification

E. Goumehei, V. Tolpekin, A. Stein, W. Yan

Research output: Contribution to journalArticle

Abstract

Detection of surface water from satellite images is important for water management purposes like for mapping flood extents, inundation dynamics, and water resources distributions. In this research, we introduce a supervised contextual classification model to detect surface water bodies from polarimetric Synthetic Aperture Radar (SAR) data. A complex Wishart Markov Random Field (WMRF) combines Markov Random Fields with the complex Wishart distribution. It is applied on Single Look Complex Sentinel 1 data. Using Markov Random Fields, we utilize the geometry of surface water to remove speckle from SAR images. Results were compared with the Wishart Maximum Likelihood Classification (WMLC), the Gaussian Maximum Likelihood Classification, and a median filter followed by thresholding. Experiments demonstrate that the statistical representation of data using the Wishart distribution improves the F-score to 0.95 for WMRF, while it is 0.67, 0.88, and 0.91 for Gaussian Maximum Likelihood Classification, WMLC, and thresholding, respectively. The main improvement in the precision increases from 0.80 and 0.86 for WMLC and thresholding to 0.96 for WMRF. The WMRF model accurately distinguishes classes that have a similar backscatter, like water and bare soil. Hence, the high accuracy of the proposed WMRF model is a result of its robustness for water detection from Single Look Complex data. We conclude that the proposed model is a great improvement on existing methods for the detection of calm surface water bodies.

Original languageEnglish
Pages (from-to)7047-7059
Number of pages13
JournalWater Resources Research
Volume55
Issue number8
DOIs
Publication statusPublished - 2019 Aug 1

Fingerprint

synthetic aperture radar
surface water
speckle
image classification
bare soil
backscatter
water management
water resource
water body
detection
filter
geometry
water
distribution
experiment

Keywords

  • complex Wishart Markov Random Fields (WMRF)
  • Single Look Complex (SLC) SAR images
  • supervised contextual classification model
  • surface water detection

ASJC Scopus subject areas

  • Water Science and Technology

Cite this

Surface Water Body Detection in Polarimetric SAR Data Using Contextual Complex Wishart Classification. / Goumehei, E.; Tolpekin, V.; Stein, A.; Yan, W.

In: Water Resources Research, Vol. 55, No. 8, 01.08.2019, p. 7047-7059.

Research output: Contribution to journalArticle

@article{d6674710e261442684bebc619fff7d15,
title = "Surface Water Body Detection in Polarimetric SAR Data Using Contextual Complex Wishart Classification",
abstract = "Detection of surface water from satellite images is important for water management purposes like for mapping flood extents, inundation dynamics, and water resources distributions. In this research, we introduce a supervised contextual classification model to detect surface water bodies from polarimetric Synthetic Aperture Radar (SAR) data. A complex Wishart Markov Random Field (WMRF) combines Markov Random Fields with the complex Wishart distribution. It is applied on Single Look Complex Sentinel 1 data. Using Markov Random Fields, we utilize the geometry of surface water to remove speckle from SAR images. Results were compared with the Wishart Maximum Likelihood Classification (WMLC), the Gaussian Maximum Likelihood Classification, and a median filter followed by thresholding. Experiments demonstrate that the statistical representation of data using the Wishart distribution improves the F-score to 0.95 for WMRF, while it is 0.67, 0.88, and 0.91 for Gaussian Maximum Likelihood Classification, WMLC, and thresholding, respectively. The main improvement in the precision increases from 0.80 and 0.86 for WMLC and thresholding to 0.96 for WMRF. The WMRF model accurately distinguishes classes that have a similar backscatter, like water and bare soil. Hence, the high accuracy of the proposed WMRF model is a result of its robustness for water detection from Single Look Complex data. We conclude that the proposed model is a great improvement on existing methods for the detection of calm surface water bodies.",
keywords = "complex Wishart Markov Random Fields (WMRF), Single Look Complex (SLC) SAR images, supervised contextual classification model, surface water detection",
author = "E. Goumehei and V. Tolpekin and A. Stein and W. Yan",
year = "2019",
month = "8",
day = "1",
doi = "10.1029/2019WR025192",
language = "English",
volume = "55",
pages = "7047--7059",
journal = "Water Resources Research",
issn = "0043-1397",
publisher = "American Geophysical Union",
number = "8",

}

TY - JOUR

T1 - Surface Water Body Detection in Polarimetric SAR Data Using Contextual Complex Wishart Classification

AU - Goumehei, E.

AU - Tolpekin, V.

AU - Stein, A.

AU - Yan, W.

PY - 2019/8/1

Y1 - 2019/8/1

N2 - Detection of surface water from satellite images is important for water management purposes like for mapping flood extents, inundation dynamics, and water resources distributions. In this research, we introduce a supervised contextual classification model to detect surface water bodies from polarimetric Synthetic Aperture Radar (SAR) data. A complex Wishart Markov Random Field (WMRF) combines Markov Random Fields with the complex Wishart distribution. It is applied on Single Look Complex Sentinel 1 data. Using Markov Random Fields, we utilize the geometry of surface water to remove speckle from SAR images. Results were compared with the Wishart Maximum Likelihood Classification (WMLC), the Gaussian Maximum Likelihood Classification, and a median filter followed by thresholding. Experiments demonstrate that the statistical representation of data using the Wishart distribution improves the F-score to 0.95 for WMRF, while it is 0.67, 0.88, and 0.91 for Gaussian Maximum Likelihood Classification, WMLC, and thresholding, respectively. The main improvement in the precision increases from 0.80 and 0.86 for WMLC and thresholding to 0.96 for WMRF. The WMRF model accurately distinguishes classes that have a similar backscatter, like water and bare soil. Hence, the high accuracy of the proposed WMRF model is a result of its robustness for water detection from Single Look Complex data. We conclude that the proposed model is a great improvement on existing methods for the detection of calm surface water bodies.

AB - Detection of surface water from satellite images is important for water management purposes like for mapping flood extents, inundation dynamics, and water resources distributions. In this research, we introduce a supervised contextual classification model to detect surface water bodies from polarimetric Synthetic Aperture Radar (SAR) data. A complex Wishart Markov Random Field (WMRF) combines Markov Random Fields with the complex Wishart distribution. It is applied on Single Look Complex Sentinel 1 data. Using Markov Random Fields, we utilize the geometry of surface water to remove speckle from SAR images. Results were compared with the Wishart Maximum Likelihood Classification (WMLC), the Gaussian Maximum Likelihood Classification, and a median filter followed by thresholding. Experiments demonstrate that the statistical representation of data using the Wishart distribution improves the F-score to 0.95 for WMRF, while it is 0.67, 0.88, and 0.91 for Gaussian Maximum Likelihood Classification, WMLC, and thresholding, respectively. The main improvement in the precision increases from 0.80 and 0.86 for WMLC and thresholding to 0.96 for WMRF. The WMRF model accurately distinguishes classes that have a similar backscatter, like water and bare soil. Hence, the high accuracy of the proposed WMRF model is a result of its robustness for water detection from Single Look Complex data. We conclude that the proposed model is a great improvement on existing methods for the detection of calm surface water bodies.

KW - complex Wishart Markov Random Fields (WMRF)

KW - Single Look Complex (SLC) SAR images

KW - supervised contextual classification model

KW - surface water detection

UR - http://www.scopus.com/inward/record.url?scp=85070881867&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070881867&partnerID=8YFLogxK

U2 - 10.1029/2019WR025192

DO - 10.1029/2019WR025192

M3 - Article

AN - SCOPUS:85070881867

VL - 55

SP - 7047

EP - 7059

JO - Water Resources Research

JF - Water Resources Research

SN - 0043-1397

IS - 8

ER -