Bayesian image reconstruction for improving detection performance of muon tomography

Guobao Wang, Larry J. Schultz, Jinyi Qi

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

Muon tomography is a novel technology that is being developed for detecting high-Z materials in vehicles or cargo containers. Maximum likelihood methods have been developed for reconstructing the scattering density image from muon measurements. However, the instability of maximum likelihood estimation often results in noisy images and low detectability of high-Z targets. In this paper, we propose using regularization to improve the image quality of muon tomography. We formulate the muon reconstruction problem in a Bayesian framework by introducing a prior distribution on scattering density images. An iterative shrinkage algorithm is derived to maximize the log posterior distribution. At each iteration, the algorithm obtains the maximum a posteriori update by shrinking an unregularized maximum likelihood update. Inverse quadratic shrinkage functions are derived for generalized Laplacian priors and inverse cubic shrinkage functions are derived for generalized Gaussian priors. Receiver operating characteristic studies using simulated data demonstrate that the Bayesian reconstruction can greatly improve the detection performance of muon tomography.

Original languageEnglish (US)
Pages (from-to)1080-1089
Number of pages10
JournalIEEE Transactions on Image Processing
Volume18
Issue number5
DOIs
StatePublished - 2009

Fingerprint

Image reconstruction
Tomography
Maximum likelihood
Scattering
Maximum likelihood estimation
Image quality
Containers

Keywords

  • Bayesian estimation
  • Expectation maximization
  • Image reconstruction
  • Muon tomography
  • ROC analysis
  • Shrinkage algorithm

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

Bayesian image reconstruction for improving detection performance of muon tomography. / Wang, Guobao; Schultz, Larry J.; Qi, Jinyi.

In: IEEE Transactions on Image Processing, Vol. 18, No. 5, 2009, p. 1080-1089.

Research output: Contribution to journalArticle

@article{90c4c455a83d4256b23ef757102fb2aa,
title = "Bayesian image reconstruction for improving detection performance of muon tomography",
abstract = "Muon tomography is a novel technology that is being developed for detecting high-Z materials in vehicles or cargo containers. Maximum likelihood methods have been developed for reconstructing the scattering density image from muon measurements. However, the instability of maximum likelihood estimation often results in noisy images and low detectability of high-Z targets. In this paper, we propose using regularization to improve the image quality of muon tomography. We formulate the muon reconstruction problem in a Bayesian framework by introducing a prior distribution on scattering density images. An iterative shrinkage algorithm is derived to maximize the log posterior distribution. At each iteration, the algorithm obtains the maximum a posteriori update by shrinking an unregularized maximum likelihood update. Inverse quadratic shrinkage functions are derived for generalized Laplacian priors and inverse cubic shrinkage functions are derived for generalized Gaussian priors. Receiver operating characteristic studies using simulated data demonstrate that the Bayesian reconstruction can greatly improve the detection performance of muon tomography.",
keywords = "Bayesian estimation, Expectation maximization, Image reconstruction, Muon tomography, ROC analysis, Shrinkage algorithm",
author = "Guobao Wang and Schultz, {Larry J.} and Jinyi Qi",
year = "2009",
doi = "10.1109/TIP.2009.2014423",
language = "English (US)",
volume = "18",
pages = "1080--1089",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Bayesian image reconstruction for improving detection performance of muon tomography

AU - Wang, Guobao

AU - Schultz, Larry J.

AU - Qi, Jinyi

PY - 2009

Y1 - 2009

N2 - Muon tomography is a novel technology that is being developed for detecting high-Z materials in vehicles or cargo containers. Maximum likelihood methods have been developed for reconstructing the scattering density image from muon measurements. However, the instability of maximum likelihood estimation often results in noisy images and low detectability of high-Z targets. In this paper, we propose using regularization to improve the image quality of muon tomography. We formulate the muon reconstruction problem in a Bayesian framework by introducing a prior distribution on scattering density images. An iterative shrinkage algorithm is derived to maximize the log posterior distribution. At each iteration, the algorithm obtains the maximum a posteriori update by shrinking an unregularized maximum likelihood update. Inverse quadratic shrinkage functions are derived for generalized Laplacian priors and inverse cubic shrinkage functions are derived for generalized Gaussian priors. Receiver operating characteristic studies using simulated data demonstrate that the Bayesian reconstruction can greatly improve the detection performance of muon tomography.

AB - Muon tomography is a novel technology that is being developed for detecting high-Z materials in vehicles or cargo containers. Maximum likelihood methods have been developed for reconstructing the scattering density image from muon measurements. However, the instability of maximum likelihood estimation often results in noisy images and low detectability of high-Z targets. In this paper, we propose using regularization to improve the image quality of muon tomography. We formulate the muon reconstruction problem in a Bayesian framework by introducing a prior distribution on scattering density images. An iterative shrinkage algorithm is derived to maximize the log posterior distribution. At each iteration, the algorithm obtains the maximum a posteriori update by shrinking an unregularized maximum likelihood update. Inverse quadratic shrinkage functions are derived for generalized Laplacian priors and inverse cubic shrinkage functions are derived for generalized Gaussian priors. Receiver operating characteristic studies using simulated data demonstrate that the Bayesian reconstruction can greatly improve the detection performance of muon tomography.

KW - Bayesian estimation

KW - Expectation maximization

KW - Image reconstruction

KW - Muon tomography

KW - ROC analysis

KW - Shrinkage algorithm

UR - http://www.scopus.com/inward/record.url?scp=65149105188&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=65149105188&partnerID=8YFLogxK

U2 - 10.1109/TIP.2009.2014423

DO - 10.1109/TIP.2009.2014423

M3 - Article

C2 - 19342340

AN - SCOPUS:65149105188

VL - 18

SP - 1080

EP - 1089

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 5

ER -