Statistical image reconstruction for muon tomography using gaussian scale mixture model

Guobao Wang, Jinyi Qi

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

3 Citations (Scopus)

Abstract

Muon tomographyis a novel imaging technique that uses background cosmic radiation to inspect cargo containers for detecting the transportation or smuggling of heavy nuclear materials. Empirically, muon scattering data are modeled as zero-mean Gaussian random variables with variance being a function of the atom number and density of the scattering material. However, a single Gaussian distribution cannot model the tail of the true distribution and hence results in inaccuracy in the reconstructed images. In this paper, we propose a Gaussian scale mixture (GSM) to approximate the true distribution of muon data. The GSM follows the true distribution more closely than a single Gaussian model. We have derived a maximum likelihood reconstruction algorithm using the optimization transfer principle. Receiver operating characteristics (ROC) studies were performed using computer simulated data to evaluate the new algorithm. The results show that the use of GSM improves the detection performance significantly over that of the traditional Gaussian model.

Original languageEnglish (US)
Title of host publicationProceedings - International Conference on Image Processing, ICIP
Pages2948-2951
Number of pages4
DOIs
StatePublished - 2008
Event2008 IEEE International Conference on Image Processing, ICIP 2008 - San Diego, CA, United States
Duration: Oct 12 2008Oct 15 2008

Other

Other2008 IEEE International Conference on Image Processing, ICIP 2008
CountryUnited States
CitySan Diego, CA
Period10/12/0810/15/08

Fingerprint

Image reconstruction
Tomography
Scattering
Cosmic rays
Gaussian distribution
Random variables
Maximum likelihood
Containers
Imaging techniques
Atoms

Keywords

  • Gaussian scale mixture
  • Image reconstruction
  • Muon tomography

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Wang, G., & Qi, J. (2008). Statistical image reconstruction for muon tomography using gaussian scale mixture model. In Proceedings - International Conference on Image Processing, ICIP (pp. 2948-2951). [4712413] https://doi.org/10.1109/ICIP.2008.4712413

Statistical image reconstruction for muon tomography using gaussian scale mixture model. / Wang, Guobao; Qi, Jinyi.

Proceedings - International Conference on Image Processing, ICIP. 2008. p. 2948-2951 4712413.

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

Wang, G & Qi, J 2008, Statistical image reconstruction for muon tomography using gaussian scale mixture model. in Proceedings - International Conference on Image Processing, ICIP., 4712413, pp. 2948-2951, 2008 IEEE International Conference on Image Processing, ICIP 2008, San Diego, CA, United States, 10/12/08. https://doi.org/10.1109/ICIP.2008.4712413
Wang G, Qi J. Statistical image reconstruction for muon tomography using gaussian scale mixture model. In Proceedings - International Conference on Image Processing, ICIP. 2008. p. 2948-2951. 4712413 https://doi.org/10.1109/ICIP.2008.4712413
Wang, Guobao ; Qi, Jinyi. / Statistical image reconstruction for muon tomography using gaussian scale mixture model. Proceedings - International Conference on Image Processing, ICIP. 2008. pp. 2948-2951
@inproceedings{e1cd39efa57d4288be7d706524ae5bb2,
title = "Statistical image reconstruction for muon tomography using gaussian scale mixture model",
abstract = "Muon tomographyis a novel imaging technique that uses background cosmic radiation to inspect cargo containers for detecting the transportation or smuggling of heavy nuclear materials. Empirically, muon scattering data are modeled as zero-mean Gaussian random variables with variance being a function of the atom number and density of the scattering material. However, a single Gaussian distribution cannot model the tail of the true distribution and hence results in inaccuracy in the reconstructed images. In this paper, we propose a Gaussian scale mixture (GSM) to approximate the true distribution of muon data. The GSM follows the true distribution more closely than a single Gaussian model. We have derived a maximum likelihood reconstruction algorithm using the optimization transfer principle. Receiver operating characteristics (ROC) studies were performed using computer simulated data to evaluate the new algorithm. The results show that the use of GSM improves the detection performance significantly over that of the traditional Gaussian model.",
keywords = "Gaussian scale mixture, Image reconstruction, Muon tomography",
author = "Guobao Wang and Jinyi Qi",
year = "2008",
doi = "10.1109/ICIP.2008.4712413",
language = "English (US)",
isbn = "1424417643",
pages = "2948--2951",
booktitle = "Proceedings - International Conference on Image Processing, ICIP",

}

TY - GEN

T1 - Statistical image reconstruction for muon tomography using gaussian scale mixture model

AU - Wang, Guobao

AU - Qi, Jinyi

PY - 2008

Y1 - 2008

N2 - Muon tomographyis a novel imaging technique that uses background cosmic radiation to inspect cargo containers for detecting the transportation or smuggling of heavy nuclear materials. Empirically, muon scattering data are modeled as zero-mean Gaussian random variables with variance being a function of the atom number and density of the scattering material. However, a single Gaussian distribution cannot model the tail of the true distribution and hence results in inaccuracy in the reconstructed images. In this paper, we propose a Gaussian scale mixture (GSM) to approximate the true distribution of muon data. The GSM follows the true distribution more closely than a single Gaussian model. We have derived a maximum likelihood reconstruction algorithm using the optimization transfer principle. Receiver operating characteristics (ROC) studies were performed using computer simulated data to evaluate the new algorithm. The results show that the use of GSM improves the detection performance significantly over that of the traditional Gaussian model.

AB - Muon tomographyis a novel imaging technique that uses background cosmic radiation to inspect cargo containers for detecting the transportation or smuggling of heavy nuclear materials. Empirically, muon scattering data are modeled as zero-mean Gaussian random variables with variance being a function of the atom number and density of the scattering material. However, a single Gaussian distribution cannot model the tail of the true distribution and hence results in inaccuracy in the reconstructed images. In this paper, we propose a Gaussian scale mixture (GSM) to approximate the true distribution of muon data. The GSM follows the true distribution more closely than a single Gaussian model. We have derived a maximum likelihood reconstruction algorithm using the optimization transfer principle. Receiver operating characteristics (ROC) studies were performed using computer simulated data to evaluate the new algorithm. The results show that the use of GSM improves the detection performance significantly over that of the traditional Gaussian model.

KW - Gaussian scale mixture

KW - Image reconstruction

KW - Muon tomography

UR - http://www.scopus.com/inward/record.url?scp=69949173573&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69949173573&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2008.4712413

DO - 10.1109/ICIP.2008.4712413

M3 - Conference contribution

AN - SCOPUS:69949173573

SN - 1424417643

SN - 9781424417643

SP - 2948

EP - 2951

BT - Proceedings - International Conference on Image Processing, ICIP

ER -