skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Advanced algorithms for radiographic material discrimination and inspection system design

Abstract

X-ray and neutron radiography are powerful tools for non-invasively inspecting the interior of objects. Materials can be discriminated by noting how the radiographic signal changes with variations in the input spectrum or inspection mode. However, current methods are limited in their ability to differentiate when multiple materials are present, especially within large and complex objects. With X-ray radiography, the inability to distinguish materials of a similar atomic number is especially problematic. To overcome these critical limitations, we augmented our existing inverse problem framework with two important expansions: 1) adapting the previous methodology for use with multi-modal radiography and energy-integrating detectors, and 2) applying the Cramer-Rao lower bound to select an optimal set of inspection modes for a given application a priori. Adding these expanded capabilities to our algorithmic framework with adaptive regularization, we observed improved discrimination between high-Z materials, specifically plutonium and tungsten. The combined system can estimate plutonium mass within our simulated system to within 1%. Three types of inspection modes were modeled: multi-endpoint X-ray radiography alone; in combination with neutron radiography using deuterium-deuterium (DD); or in combination with neutron radiography using deuterium-tritium (DT) sources.

Authors:
; ;
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1333437
Report Number(s):
PNNL-SA-113690
Journal ID: ISSN 0168-583X; DN4001010
DOE Contract Number:  
AC05-76RL01830
Resource Type:
Journal Article
Journal Name:
Nuclear Instruments and Methods in Physics Research. Section B, Beam Interactions with Materials and Atoms
Additional Journal Information:
Journal Volume: 385; Journal ID: ISSN 0168-583X
Publisher:
Elsevier
Country of Publication:
United States
Language:
English
Subject:
X-ray radiography; neutron radiography; material discrimination; inverse problems

Citation Formats

Gilbert, Andrew J., McDonald, Benjamin S., and Deinert, Mark R. Advanced algorithms for radiographic material discrimination and inspection system design. United States: N. p., 2016. Web. doi:10.1016/j.nimb.2016.07.013.
Gilbert, Andrew J., McDonald, Benjamin S., & Deinert, Mark R. Advanced algorithms for radiographic material discrimination and inspection system design. United States. doi:10.1016/j.nimb.2016.07.013.
Gilbert, Andrew J., McDonald, Benjamin S., and Deinert, Mark R. Sat . "Advanced algorithms for radiographic material discrimination and inspection system design". United States. doi:10.1016/j.nimb.2016.07.013.
@article{osti_1333437,
title = {Advanced algorithms for radiographic material discrimination and inspection system design},
author = {Gilbert, Andrew J. and McDonald, Benjamin S. and Deinert, Mark R.},
abstractNote = {X-ray and neutron radiography are powerful tools for non-invasively inspecting the interior of objects. Materials can be discriminated by noting how the radiographic signal changes with variations in the input spectrum or inspection mode. However, current methods are limited in their ability to differentiate when multiple materials are present, especially within large and complex objects. With X-ray radiography, the inability to distinguish materials of a similar atomic number is especially problematic. To overcome these critical limitations, we augmented our existing inverse problem framework with two important expansions: 1) adapting the previous methodology for use with multi-modal radiography and energy-integrating detectors, and 2) applying the Cramer-Rao lower bound to select an optimal set of inspection modes for a given application a priori. Adding these expanded capabilities to our algorithmic framework with adaptive regularization, we observed improved discrimination between high-Z materials, specifically plutonium and tungsten. The combined system can estimate plutonium mass within our simulated system to within 1%. Three types of inspection modes were modeled: multi-endpoint X-ray radiography alone; in combination with neutron radiography using deuterium-deuterium (DD); or in combination with neutron radiography using deuterium-tritium (DT) sources.},
doi = {10.1016/j.nimb.2016.07.013},
journal = {Nuclear Instruments and Methods in Physics Research. Section B, Beam Interactions with Materials and Atoms},
issn = {0168-583X},
number = ,
volume = 385,
place = {United States},
year = {2016},
month = {10}
}