Change detection on SAR images by a parametric estimation of the KL-divergence between Gaussian Mixture Models

Qian Xu, Lina Karam

Research output: Chapter in Book/Report/Conference proceedingConference contribution

11 Citations (Scopus)

Abstract

In the context of multi-temporal synthetic aperture radar (SAR) images for earth monitoring applications, one critical issue is the detection of changes occurring after a natural or anthropic disaster. In this paper, we propose a new similarity measure for automatic change detection using a pair of SAR images acquired at different dates. This measure is based on the evolution of the local statistics of the image between two dates. The local statistics are modeled as a Gaussian Mixture Model (GMM), which approximates the probability density function in the neighborhood of each pixel in the image. The degree of evolution of the local statistics is measured using the Kullback-Leibler (KL) divergence. One analytical expression for approximating the KL divergence between GMMs is given and is compared with the Monte Carlo sampling method. The proposed change detector is compared to the classical mean ratio detector and also to other recent model-based approaches. Tests on the real data show that our detector outperforms previously suggested methods in terms of the rate of missed detections and the total error rates.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages2109-2113
Number of pages5
DOIs
StatePublished - Oct 18 2013
Event2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Vancouver, BC, Canada
Duration: May 26 2013May 31 2013

Other

Other2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013
CountryCanada
CityVancouver, BC
Period5/26/135/31/13

Fingerprint

Synthetic aperture radar
Statistics
Detectors
Disasters
Probability density function
Pixels
Earth (planet)
Sampling
Monitoring

Keywords

  • change detection
  • Gaussian mixture models
  • Kullback-Leibler (KL) divergence
  • multitemporal synthetic aperture radar (SAR) images

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Xu, Q., & Karam, L. (2013). Change detection on SAR images by a parametric estimation of the KL-divergence between Gaussian Mixture Models. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 2109-2113). [6638026] https://doi.org/10.1109/ICASSP.2013.6638026

Change detection on SAR images by a parametric estimation of the KL-divergence between Gaussian Mixture Models. / Xu, Qian; Karam, Lina.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2013. p. 2109-2113 6638026.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Xu, Q & Karam, L 2013, Change detection on SAR images by a parametric estimation of the KL-divergence between Gaussian Mixture Models. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings., 6638026, pp. 2109-2113, 2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013, Vancouver, BC, Canada, 5/26/13. https://doi.org/10.1109/ICASSP.2013.6638026
Xu Q, Karam L. Change detection on SAR images by a parametric estimation of the KL-divergence between Gaussian Mixture Models. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2013. p. 2109-2113. 6638026 https://doi.org/10.1109/ICASSP.2013.6638026
Xu, Qian ; Karam, Lina. / Change detection on SAR images by a parametric estimation of the KL-divergence between Gaussian Mixture Models. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2013. pp. 2109-2113
@inproceedings{800cc250941847ce97af8372d83083f3,
title = "Change detection on SAR images by a parametric estimation of the KL-divergence between Gaussian Mixture Models",
abstract = "In the context of multi-temporal synthetic aperture radar (SAR) images for earth monitoring applications, one critical issue is the detection of changes occurring after a natural or anthropic disaster. In this paper, we propose a new similarity measure for automatic change detection using a pair of SAR images acquired at different dates. This measure is based on the evolution of the local statistics of the image between two dates. The local statistics are modeled as a Gaussian Mixture Model (GMM), which approximates the probability density function in the neighborhood of each pixel in the image. The degree of evolution of the local statistics is measured using the Kullback-Leibler (KL) divergence. One analytical expression for approximating the KL divergence between GMMs is given and is compared with the Monte Carlo sampling method. The proposed change detector is compared to the classical mean ratio detector and also to other recent model-based approaches. Tests on the real data show that our detector outperforms previously suggested methods in terms of the rate of missed detections and the total error rates.",
keywords = "change detection, Gaussian mixture models, Kullback-Leibler (KL) divergence, multitemporal synthetic aperture radar (SAR) images",
author = "Qian Xu and Lina Karam",
year = "2013",
month = "10",
day = "18",
doi = "10.1109/ICASSP.2013.6638026",
language = "English (US)",
isbn = "9781479903566",
pages = "2109--2113",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - Change detection on SAR images by a parametric estimation of the KL-divergence between Gaussian Mixture Models

AU - Xu, Qian

AU - Karam, Lina

PY - 2013/10/18

Y1 - 2013/10/18

N2 - In the context of multi-temporal synthetic aperture radar (SAR) images for earth monitoring applications, one critical issue is the detection of changes occurring after a natural or anthropic disaster. In this paper, we propose a new similarity measure for automatic change detection using a pair of SAR images acquired at different dates. This measure is based on the evolution of the local statistics of the image between two dates. The local statistics are modeled as a Gaussian Mixture Model (GMM), which approximates the probability density function in the neighborhood of each pixel in the image. The degree of evolution of the local statistics is measured using the Kullback-Leibler (KL) divergence. One analytical expression for approximating the KL divergence between GMMs is given and is compared with the Monte Carlo sampling method. The proposed change detector is compared to the classical mean ratio detector and also to other recent model-based approaches. Tests on the real data show that our detector outperforms previously suggested methods in terms of the rate of missed detections and the total error rates.

AB - In the context of multi-temporal synthetic aperture radar (SAR) images for earth monitoring applications, one critical issue is the detection of changes occurring after a natural or anthropic disaster. In this paper, we propose a new similarity measure for automatic change detection using a pair of SAR images acquired at different dates. This measure is based on the evolution of the local statistics of the image between two dates. The local statistics are modeled as a Gaussian Mixture Model (GMM), which approximates the probability density function in the neighborhood of each pixel in the image. The degree of evolution of the local statistics is measured using the Kullback-Leibler (KL) divergence. One analytical expression for approximating the KL divergence between GMMs is given and is compared with the Monte Carlo sampling method. The proposed change detector is compared to the classical mean ratio detector and also to other recent model-based approaches. Tests on the real data show that our detector outperforms previously suggested methods in terms of the rate of missed detections and the total error rates.

KW - change detection

KW - Gaussian mixture models

KW - Kullback-Leibler (KL) divergence

KW - multitemporal synthetic aperture radar (SAR) images

UR - http://www.scopus.com/inward/record.url?scp=84890443855&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890443855&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2013.6638026

DO - 10.1109/ICASSP.2013.6638026

M3 - Conference contribution

SN - 9781479903566

SP - 2109

EP - 2113

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -