Improved Acquisition and Reconstruction for Wavelength-Resolved Neutron Tomography

Abstract

Wavelength-resolved neutron tomography (WRNT) is an emerging technique for characterizing samples relevant to the materials sciences in 3D. WRNT studies can be carried out at beam lines in spallation neutron or reactor-based user facilities. Because of the limited availability of experimental time, potential imperfections in the neutron source, or constraints placed on the acquisition time by the type of sample, the data can be extremely noisy resulting in tomographic reconstructions with significant artifacts when standard reconstruction algorithms are used. Furthermore, making a full tomographic measurement even with a low signal-to-noise ratio can take several days, resulting in a long wait time before the user can receive feedback from the experiment when traditional acquisition protocols are used. In this paper, we propose an interlaced scanning technique and combine it with a model-based image reconstruction algorithm to produce high-quality WRNT reconstructions concurrent with the measurements being made. The interlaced scan is designed to acquire data so that successive measurements are more diverse in contrast to typical sequential scanning protocols. The model-based reconstruction algorithm combines a data-fidelity term with a regularization term to formulate the wavelength-resolved reconstruction as minimizing a high-dimensional cost-function. Using an experimental dataset of a magnetite sample acquired over a span of about two days, we demonstrate that our technique can produce high-quality reconstructions even during the experiment compared to traditional acquisition and reconstruction techniques. In summary, the combination of the proposed acquisition strategy with an advanced reconstruction algorithm provides a novel guideline for designing WRNT systems at user facilities.

Neutron Imaging and Learning Algorithms: New Perspectives in Cultural Heritage Applications

“…Energy-resolved Neutron Imaging has developed in recent years owing to the use of bright pulsed neutron beamline facilities at neutron spallation sources and is applied to study the composition, strain, and texture of the sample as a function of the spatial resolution [ 62 , 63 , 64 ]. Because the data are affected by noise, Machine Learning could be successfully applied to cultural heritage optimizing the acquisition time and improving the quantification results.…”

Algorithm-driven Advances for Scientific CT Instruments: From Model-based to Deep Learning-based Approaches

“…For example, it is common across SCT applications to use detectors which are approximately 2000 × 2000 pixels, and corresponding CT reconstructions to have sizes of the order of 2000 × 2000 × 2000 voxels. In the case of hyper-spectral SCT instruments [11],…”

“…4DCT is used to study dynamic phenomena such as solidification [41], phase transformations [42], crack formation [43], and battery degradation [44]- [46]. 4DCT is performed using a variety of radiation sources that include X-rays [44], electrons [47], and neutrons [48], [49] at resolutions ranging from nanometer to micron length scales mostly using the set-up of the type in Fig. 1 (a).…”

A machine learning decision criterion for reducing scan time for hyperspectral neutron computed tomography systems

“…In order to address the challenges of long scan times and limited user feedback, advanced scanning protocols based on the golden ratio method 29 combined with model-based iterative reconstruction (MBIR) algorithms have recently been proposed 30 by our team. It was demonstrated that by streaming data to a powerful compute node, and using a fast implementation of MBIR 31 , it was possible to reconstruct CT scans with high image quality while performing an experiment, thereby paving the way to make real-time decision during an experiment.…”