Quantcast

Advanced algorithms for radiographic material discrimination and inspection system design

Research paper by Andrew J. Gilbert, Benjamin S. McDonald; Mark R. Deinert

Indexed on: 18 Oct '16Published on: 20 Sep '16Published in: Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms



Abstract

Publication date: 15 October 2016 Source:Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, Volume 385 Author(s): Andrew J. Gilbert, Benjamin S. McDonald, Mark R. Deinert X-ray and neutron radiography are powerful tools for non-invasively inspecting the interior of objects. However, current methods are limited in their ability to differentiate materials when multiple materials are present, especially within large and complex objects. Past work has demonstrated that the spectral shift that X-ray beams undergo in traversing an object can be used to detect and quantify nuclear materials. The technique uses a spectrally sensitive detector and an inverse algorithm that varies the composition of the object until the X-ray spectrum predicted by X-ray transport matches the one measured. Here we show that this approach can be adapted to multi-mode radiography, with energy integrating detectors, and that the Cramér–Rao lower bound can be used to choose an optimal set of inspection modes a priori. We consider multi-endpoint X-ray radiography alone, or in combination with neutron radiography using deuterium–deuterium (DD) or deuterium–tritium (DT) sources. We show that for an optimal mode choice, the algorithm can improve discrimination between high-Z materials, specifically between tungsten and plutonium, and estimate plutonium mass within a simulated nuclear material storage system to within 1%.