Parameter estimation of heavy-tailed rayleigh prior model from observed SAR amplitude images

Zengguo Sun, Ram Mohan Narayanan, Chongzhao Han

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

1 Citation (Scopus)

Abstract

We propose a new method for estimating the parameters of heavy-tailed Rayleigh prior model from the observed synthetic aperture radar (SAR) amplitude images. First, we change the multiplicative SAR image model into an additive one using logarithmic transformation. Next, we derive the expectation and variance of the log-transformed image using negative-order moments concept. Finally, we obtain the above quantities for two kinds of SAR amplitude images: the squareroot of intensity image and the multi-look averaged amplitude image. Particularly for the latter, we derive the closed-form expressions for such expectation and variance based on the Gaussian approximation of the log-transformed speckle. Monte Carlo simulation results demonstrate that the proposed estimators, which are easy to implement with the analytical expressions, are efficient for the parameter estimation of the heavy-tailed Rayleigh prior model from the observed image.

Original languageEnglish (US)
Title of host publication2008 5th European Radar Conference Proceedings, EuRAD 2008
Pages360-363
Number of pages4
StatePublished - 2008
Event2008 5th European Radar Conference, EuRAD 2008 - Amsterdam, Netherlands
Duration: Oct 30 2008Oct 31 2008

Other

Other2008 5th European Radar Conference, EuRAD 2008
CountryNetherlands
CityAmsterdam
Period10/30/0810/31/08

Fingerprint

Synthetic aperture radar
Parameter estimation
Speckle

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Sun, Z., Narayanan, R. M., & Han, C. (2008). Parameter estimation of heavy-tailed rayleigh prior model from observed SAR amplitude images. In 2008 5th European Radar Conference Proceedings, EuRAD 2008 (pp. 360-363). [4760876]
Sun, Zengguo ; Narayanan, Ram Mohan ; Han, Chongzhao. / Parameter estimation of heavy-tailed rayleigh prior model from observed SAR amplitude images. 2008 5th European Radar Conference Proceedings, EuRAD 2008. 2008. pp. 360-363
@inproceedings{19f4150f4448471ba701314bab7c77d6,
title = "Parameter estimation of heavy-tailed rayleigh prior model from observed SAR amplitude images",
abstract = "We propose a new method for estimating the parameters of heavy-tailed Rayleigh prior model from the observed synthetic aperture radar (SAR) amplitude images. First, we change the multiplicative SAR image model into an additive one using logarithmic transformation. Next, we derive the expectation and variance of the log-transformed image using negative-order moments concept. Finally, we obtain the above quantities for two kinds of SAR amplitude images: the squareroot of intensity image and the multi-look averaged amplitude image. Particularly for the latter, we derive the closed-form expressions for such expectation and variance based on the Gaussian approximation of the log-transformed speckle. Monte Carlo simulation results demonstrate that the proposed estimators, which are easy to implement with the analytical expressions, are efficient for the parameter estimation of the heavy-tailed Rayleigh prior model from the observed image.",
author = "Zengguo Sun and Narayanan, {Ram Mohan} and Chongzhao Han",
year = "2008",
language = "English (US)",
isbn = "9782874870095",
pages = "360--363",
booktitle = "2008 5th European Radar Conference Proceedings, EuRAD 2008",

}

Sun, Z, Narayanan, RM & Han, C 2008, Parameter estimation of heavy-tailed rayleigh prior model from observed SAR amplitude images. in 2008 5th European Radar Conference Proceedings, EuRAD 2008., 4760876, pp. 360-363, 2008 5th European Radar Conference, EuRAD 2008, Amsterdam, Netherlands, 10/30/08.

Parameter estimation of heavy-tailed rayleigh prior model from observed SAR amplitude images. / Sun, Zengguo; Narayanan, Ram Mohan; Han, Chongzhao.

2008 5th European Radar Conference Proceedings, EuRAD 2008. 2008. p. 360-363 4760876.

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

TY - GEN

T1 - Parameter estimation of heavy-tailed rayleigh prior model from observed SAR amplitude images

AU - Sun, Zengguo

AU - Narayanan, Ram Mohan

AU - Han, Chongzhao

PY - 2008

Y1 - 2008

N2 - We propose a new method for estimating the parameters of heavy-tailed Rayleigh prior model from the observed synthetic aperture radar (SAR) amplitude images. First, we change the multiplicative SAR image model into an additive one using logarithmic transformation. Next, we derive the expectation and variance of the log-transformed image using negative-order moments concept. Finally, we obtain the above quantities for two kinds of SAR amplitude images: the squareroot of intensity image and the multi-look averaged amplitude image. Particularly for the latter, we derive the closed-form expressions for such expectation and variance based on the Gaussian approximation of the log-transformed speckle. Monte Carlo simulation results demonstrate that the proposed estimators, which are easy to implement with the analytical expressions, are efficient for the parameter estimation of the heavy-tailed Rayleigh prior model from the observed image.

AB - We propose a new method for estimating the parameters of heavy-tailed Rayleigh prior model from the observed synthetic aperture radar (SAR) amplitude images. First, we change the multiplicative SAR image model into an additive one using logarithmic transformation. Next, we derive the expectation and variance of the log-transformed image using negative-order moments concept. Finally, we obtain the above quantities for two kinds of SAR amplitude images: the squareroot of intensity image and the multi-look averaged amplitude image. Particularly for the latter, we derive the closed-form expressions for such expectation and variance based on the Gaussian approximation of the log-transformed speckle. Monte Carlo simulation results demonstrate that the proposed estimators, which are easy to implement with the analytical expressions, are efficient for the parameter estimation of the heavy-tailed Rayleigh prior model from the observed image.

UR - http://www.scopus.com/inward/record.url?scp=63149125561&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=63149125561&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9782874870095

SP - 360

EP - 363

BT - 2008 5th European Radar Conference Proceedings, EuRAD 2008

ER -

Sun Z, Narayanan RM, Han C. Parameter estimation of heavy-tailed rayleigh prior model from observed SAR amplitude images. In 2008 5th European Radar Conference Proceedings, EuRAD 2008. 2008. p. 360-363. 4760876