The time-of-flight imaging beamline VENUS at the Spallation Neutron Source of the Oak Ridge National Laboratory has recently entered the commissioning phase. Based on the experience acquired at the Spallation Neutron Source SNAP and the High Flux Isotope Reactor MARS beamlines, we have designed various analysis tools and a new intuitive data acquisition system user interface. VENUS starts commissioning with the micro-channel plate Timepix detector and light sensitive cameras that include a charge-coupled device and complementary metal oxide semiconductor sensors. A micro-channel plate (MCP) Timepix3 detector is currently being tested at VENUS and is in the process of being integrated to the VENUS data acquisition system. Since this new detector is capable of measuring data in event mode, both hardware and data workflow require careful planning as described in this manuscript. A new version of the in-house Imaging Bragg Edge Analysis Tools for Engineering Structures (iBeatles) has been produced thanks to the feedback received at the 11th workshop on neutron wavelength-dependent imaging held in Japan in 2023. This paper gives an overview of the VENUS hardware and software infrastructures, the data workflow, the VENUS data acquisition interface and features available in the new version of iBeatles.

Industrial X-ray computed tomography (XCT) enables non-destructive inspection of additively manufactured (AM) parts, but high-resolution scanning requires long acquisition times and significant computational resources, limiting throughput in production environments. Super-resolution techniques can recover high-resolution information from low-resolution scans, but existing methods face a trade-off between 2D approaches that ignore inter-slice information and 3D methods that are computationally prohibitive for practical deployment. To address this trade-off, we propose a 2.5D deep learning-based super-resolution approach that uses seven neighbouring low-resolution slices to super-resolve the centre slice. This work evaluates the method on real XCT scans of steel AM parts, comparing reconstruction quality and flaw detection performance of 2D, 2.5D, and 3D ESRGAN-based super-resolution methods. Results demonstrate that 2.5D super-resolution significantly improves detection of small, process-induced flaws (e.g. porosity) compared to 2D methods, while avoiding the prohibitive computational burden of full 3D approaches. These findings provide initial evidence of 2.5D super-resolution as a practical, deployable solution for improving flaw detection in high-throughput industrial XCT inspection.

Hyperspectral neutron computed tomography enables 3D non-destructive imaging of the spectral characteristics of materials. In traditional hyperspectral reconstruction, the data for each neutron wavelength bin is reconstructed separately. This per-bin reconstruction is extremely time-consuming due to the typically large number of wavelength bins. Furthermore, these reconstructions may suffer from severe artifacts due to the low signal-to-noise ratio in each wavelength bin. We present a novel fast hyperspectral reconstruction algorithm for computationally efficient and accurate reconstruction of hyperspectral neutron data. Our algorithm uses a subspace extraction procedure that transforms hyperspectral data into low-dimensional data within an intermediate subspace. This step effectively reduces data dimensionality and spectral noise. High-quality reconstructions are then performed within this low-dimensional subspace. Finally, the algorithm expands the subspace reconstructions into hyperspectral reconstructions. We apply our algorithm to measured neutron data and demonstrate that it reduces computation and improves reconstruction quality compared to the conventional approach.

Diffusion models have emerged as powerful priors for solving inverse problems in computed tomography (CT). In certain applications, such as neutron CT, it can be expensive to collect large amounts of measurements even for a single scan, leading to sparse data sets from which it is challenging to obtain high quality reconstructions even with diffusion models. One strategy to mitigate this challenge is to leverage a complementary, easily available imaging modality; however, such approaches typically require retraining the diffusion model with large datasets. In this work, we propose incorporating an additional modality without retraining the diffusion prior, enabling accelerated imaging of costly modalities. We further examine the impact of imperfect side modalities on cross-modal guidance. Our method is evaluated on sparse-view neutron computed tomography, where reconstruction quality is substantially improved by incorporating X-ray computed tomography of the same samples.

Cone-beam X-ray Computed Tomography (XCT) with large detectors and corresponding large-scale 3D reconstruction plays a pivotal role in micron-scale characterization of materials and parts across various industries. In this work, we present a novel deep neural network-based iterative algorithm that integrates an artifact reduction-trained CNN as a prior model with automated regularization parameter selection, tailored for large-scale industrial cone-beam XCT data. Our method achieves high-quality 3D reconstructions even for extremely dense thick metal parts - which traditionally pose challenges to industrial CT images - in just a few iterations. Furthermore, we show the generalizability of our approach to out-of-distribution scans obtained under diverse scanning conditions. Our method effectively handles significant noise and streak artifacts, surpassing state-of-the-art supervised learning methods trained on the same data.

Industrial cone-beam X-ray computed tomography (CT) produces 3D reconstructions of objects using projection measurements taken at multiple predetermined rotation angles around a single axis. Achieving high-quality reconstructions with traditional analytic reconstruction algorithms typically requires a large number of projections, which can be both time-consuming and costly. State-of-the-art reconstruction algorithms, such as model-based iterative reconstruction (MBIR), have made it possible to achieve high-quality reconstructions using significantly fewer projections. However, the process of acquiring these sparse projections often fails to account for the specific geometry of the scanned object. In this paper, we propose an algorithm to optimize the scanning process by using geometric information about the object to be scanned. Our approach strategically selects the most informative views by assessing their alignment with the object’s long edges while ensuring that the selected projections maintain sufficient diversity. To make the algorithm practical for large 3D volumes, we developed a two-stage method. The first stage uses a low-resolution version of the voxelized CAD model to obtain edge information and combines it with view diversity information for the optimization in the initial iterations, while the second stage relies on a single low-resolution reconstruction based on measurements from the first stage to gather edge and view diversity information during the later iterations. During the view selection process, all reconstructions are performed using MBIRJAX, a novel software library that enables fast, high-quality cone-beam CT reconstructions. Through simulations and measured datasets, we demonstrate that our algorithm produces higher quality reconstructions, particularly in preserving sharp edges, while requiring fewer measurements compared to the traditional method.

Neutron imaging is a non-destructive technique used to study the internal structure and composition of a wide range of materials. At Oak Ridge National Laboratory (ORNL), there are two neutron imaging beamlines that provide complementary capabilities that serve a diverse community, from materials science and engineering to biomedical and plant sciences. This report provides an overview of the current ORNL imaging capabilities that support materials research for a broad user community.

X-ray computed tomography (XCT) is essential for nondestructive evaluation and quality control of large-scale metal components. XCT imaging, however, faces significant challenges from metal artifacts, particularly those caused by Compton scattering, which degrade image quality and obscure critical details. Hardware-based solutions (e.g. scatterControl) offer advancements by intercepting scattered photons and reducing artifacts, but they can be time-consuming and require additional processing. Here, we propose modifying and leveraging a novel deep learning (DL) framework, Simurgh, to enhance and accelerate scatter correction in XCT. By combining scatterControl with DL-based artifact removal, we demonstrate significant reduction in scan time while producing high-quality reconstructions. Through extensive evaluation on industrial XCT data, we show that our methods reduce scan time by up to more than 10 × while preserving flaw detectability. Quantitative analysis across multiple segmentation techniques confirms that Simurgh-based reconstructions consistently outperform traditional Feldkamp-Davis-Kress, model-based iterative reconstruction, and commercial DL models in both pixel-level and task-specific evaluations, enabling scalable, high-throughput XCT workflows for characterization of large scale components in applications such as casting and metal additive manufacturing.

Industrial X-ray cone-beam CT (XCT) scanners are widely used for scientific imaging and non-destructive characterization. Industrial CBCT scanners use large detectors containing millions of pixels and the subsequent 3D reconstructions can be of the order of billions of voxels. In order to obtain high-quality reconstruction when using typical analytic algorithms, the scan involves collecting a large number of projections/views which results in large measurement times - limiting the utility of the technique. Model-based iterative reconstruction (MBIR) algorithms can produce high-quality reconstructions from fast sparse-view CT scans, but are computationally expensive and hence are avoided in practice. Single-step deep-learning (DL) based methods have demonstrated that it is possible to obtain fast and high-quality reconstructions from sparse-view data but they do not generalize well to out-of-distribution scenarios. In this work, we propose a half-quadratic splitting-based algorithm that uses convolutional neural networks (CNN) in order to obtain high-quality reconstructions from large sparse-view cone-beam CT (CBCT) measurements while overcoming the challenges with typical approaches. The algorithm alternates between the application of a CNN and a conjugate gradient (CG) step enforcing data-consistency (DC). The proposed method outperforms other methods on the publicly available Walnuts data-set.

Hyperspectralneutron computed tomography is a tomographic imaging technique in which thousands of wavelength-specific neutron radiographs are measured for each tomographic view. In conventional hyperspectral reconstruction, data from each neutron wavelength bin are reconstructed separately, which is extremely time-consuming. These reconstructions often suffer from poor quality due to low signal-to-noise ratios. Consequently, material decomposition based on these reconstructions tends to produce inaccurate estimates of the material spectra and erroneous volumetric material separation. In this paper, we present two novel algorithms for processing hyperspectral neutron data: fast hyperspectral reconstruction and fast material decomposition. Both algorithms rely on a subspace decomposition procedure that transforms hyperspectral views into low-dimensional projection views within an intermediate subspace, where tomographic reconstruction is performed. The use of subspace decomposition dramatically reduces reconstruction time while reducing both noise and reconstruction artifacts. We apply our algorithms to both simulated and measured neutron data and demonstrate that they reduce computation and improve the quality of the results relative to conventional methods.

Cone-beam X-ray computed tomography (XCT) is an essential imaging technique for generating 3D reconstructions of internal structures, with applications ranging from medical to industrial imaging. Producing high-quality reconstructions typically requires many X-ray measurements; this process can be slow and expensive, especially for dense materials. Recent work incorporating artifact reduction priors within a plug-and-play (PnP) reconstruction framework has shown promising results in improving image quality from sparse-view XCT scans while enhancing the generalizability of deep learning-based solutions. However, this method uses a 2D convolutional neural network (CNN) for artifact reduction, which captures only slice-independent information from the 3D reconstruction, limiting performance. In this paper, we propose a PnP reconstruction method that uses a 2.5D artifact reduction CNN as the prior. This approach leverages inter-slice information from adjacent slices, capturing richer spatial context while remaining computationally efficient. We show that this 2.5D prior not only improves the quality of reconstructions but also enables the model to directly suppress commonly occurring XCT artifacts (such as beam hardening), eliminating the need for artifact correction pre-processing. Experiments on both experimental and synthetic cone-beam XCT data demonstrate that the proposed method better preserves fine structural details, such as pore size and shape, leading to more accurate defect detection compared to 2D priors. In particular, we demonstrate strong performance on experimental XCT data using a 2.5D artifact reduction prior trained entirely on simulated scans, highlighting the proposed method's ability to generalize across domains.

Obtaining high-quality 3D reconstructions from electron tomography of crystalline particles embedded in lighter support elements is crucial for various material systems such as catalysts for fuel cell applications. However, significant challenges arise due to the limited tilt range, sparse and low signal-to-noise ratio of the measurements. In addition, small metal particles can cause strong streaking and shading artifacts in the 3D reconstructions when using conventional reconstruction algorithms due to the presence of Bragg diffraction and the large scattering cross-section difference between the materials of the particles and the background support regions. These artifacts lead to errors in the downstream characterization affecting extraction of critical features such as the size of the metal particles, their distribution and the volume of the lighter support regions. In this paper, we present a two-stage algorithm based on metal artifact reduction, utilizing model-based iterative reconstruction methods with adaptive adjustment of regularization parameters. Our approach yields high-quality 3D reconstructions compared to traditional algorithms, accurately capturing both the metal particles as well as the background support. We demonstrate the effectiveness of our algorithm through simulated and experimental bright-field electron tomography data, showing significant improvements in reconstruction quality compared to traditional methods.

Industrial cone-beam X-ray computed tomography (CT) uses projections acquired at multiple predetermined rotation angles about a single axis to produce a 3D reconstruction of a target object. Typically, a large number of projections are required to achieve a high-quality reconstruction, which can take several hours or days depending on the part size, material composition, and desired resolution. In this paper, we introduce an adaptive view selection procedure designed to optimize the scanning process by selecting the best next set of angles based on the current reconstruction and the computer-aided design (CAD) model. We develop a score function to identify angles aligned with long edges in the object, and we develop a tunable angle dispersion score that can be used to determine the importance of angular diversity. We combine these scores to promote measurements aligned with the part's long edges while maintaining a diverse set of overall measurements. Through simulations, we demonstrate that our algorithm notably improves reconstruction quality for a small number of views relative to traditional methods.

We present the first machine learning-based autonomous hyperspectral neutron computed tomography experiment performed at the Spallation Neutron Source. Hyperspectral neutron computed tomography allows the characterization of samples by enabling the reconstruction of crystallographic information and elemental/isotopic composition of objects relevant to materials science. High quality reconstructions using traditional algorithms such as the filtered back projection require a high signal-to-noise ratio across a wide wavelength range combined with a large number of projections. This results in scan times of several days to acquire hundreds of hyperspectral projections, during which end users have minimal feedback. To address these challenges, a golden ratio scanning protocol combined with model-based image reconstruction algorithms have been proposed. This novel approach enables high quality real-time reconstructions from streaming experimental data, thus providing feedback to users, while requiring fewer yet a fixed number of projections compared to the filtered back projection method. In this paper, we propose a novel machine learning criterion that can terminate a streaming neutron tomography scan once sufficient information is obtained based on the current set of measurements. Our decision criterion uses a quality score which combines a reference-free image quality metric computed using a pre-trained deep neural network with a metric that measures differences between consecutive reconstructions. The results show that our method can reduce the measurement time by approximately a factor of five compared to a baseline method based on filtered back projection for the samples we studied while automatically terminating the scans.

X-ray computed tomography (XCT) is a key tool in non-destructive evaluation of additively manufactured (AM) parts, allowing for internal inspection and defect detection. Despite its widespread use, obtaining high-resolution CT scans can be extremely time consuming. This issue can be mitigated by performing scans at lower resolutions; however, reducing the resolution compromises spatial detail, limiting the accuracy of defect detection. Super-resolution algorithms offer a promising solution for overcoming resolution limitations in XCT reconstructions of AM parts, enabling more accurate detection of defects. While 2D super-resolution methods have demonstrated state-of-the-art performance on natural images, they tend to under-perform when directly applied to XCT slices. On the other hand, 3D super-resolution methods are computationally expensive, making them infeasible for large-scale applications. To address these challenges, we propose a 2.5D super-resolution approach tailored for XCT of AM parts. Our method enhances the resolution of individual slices by leveraging multi-slice information from neighboring 2D slices without the significant computational overhead of full 3D methods. Specifically, we use neighboring low-resolution slices to super-resolve the center slice, exploiting inter-slice spatial context while maintaining computational efficiency. This approach bridges the gap between 2D and 3D methods, offering a practical solution for high-throughput defect detection in AM parts.

Located at the Spallation Neutron Source, VENUS is an instrument optimized for wavelength-dependent neutron imaging techniques, namely Bragg edge and resonance imaging, across a broad range of neutron energies (from meV to hundreds of eV). The VENUS construction project started in Fall 2018, establishing a conceptual design of the major components and timeline for purchasing, testing, and installation. Completion of the VENUS instrument is anticipated late 2024. This instrument comprises components such as a set of boron carbide apertures designed for both thermal/cold and epithermal neutrons, a suite of choppers (T 0 and single disks), beam collimators, and a cadmium filter. The time-of-flight detector at VENUS will be the micro-channel-plate Timepix detector and will be positioned 25 m away from the source. This manuscript describes the instrument general layout, explains the choice of the moderator, and specifies the VENUS key capabilities. It also provides the status of the construction project with installed components. Discussions about software development are also presented.

Hyperspectral neutron tomography is an effective method for analyzing crystalline material samples with complex compositions in a non-destructive manner. Since the counts in the hyperspectral neutron radiographs directly depend on the neutron cross-sections, materials may exhibit contrasting neutron responses across wavelengths. Therefore, it is possible to extract the unique signatures associated with each material and use them to separate the crystalline phases simultaneously.We introduce an autonomous material decomposition (AMD) algorithm to automatically characterize and localize polycrystalline structures using Bragg edges with contrasting neutron responses from hyperspectral data. The algorithm estimates the linear attenuation coefficient spectra from the measured radiographs and then uses these spectra to perform polycrystalline material decomposition and reconstructs 3D material volumes to localize materials in the spatial domain. Our results demonstrate that the method can accurately estimate both the linear attenuation coefficient spectra and associated reconstructions on both simulated and experimental neutron data.

We introduce a physics guided data-driven method for image-based multi-material decomposition for dual-energy computed tomography (CT) scans. The method is demonstrated for CT scans of virtual human phantoms containing more than two types of tissues. The method is a physics-driven supervised learning technique. We take advantage of the mass attenuation coefficient of dense materials compared to that of muscle tissues to perform a preliminary extraction of the dense material from the images using unsupervised methods. We then perform supervised deep learning on the images processed by the extracted dense material to obtain the final multi-material tissue map. The method is demonstrated on simulated breast models with calcifications as the dense material placed amongst the muscle tissues. The physics-guided machine learning method accurately decomposes the various tissues from input images, achieving a normalized root-mean-squared error of 2.75%.