SEARCH · Engineering Papers

Results for “Imaging”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,567 records · Page 87

All sources NASA NTRS DOE OSTI

DOE OSTI2016

TU-FG-209-11: Validation of a Channelized Hotelling Observer to Optimize Chest Radiography Image Processing for Nodule Detection: A Human Observer Study

Purpose: To validate the use of a Channelized Hotelling Observer (CHO) model for guiding image processing parameter selection and enable improved nodule detection in digital chest radiography. Methods: In a previous study, an anthropomorphic chest phantom was imaged with and without PMMA simulated nodules using a GE Discovery XR656 digital radiography system. The impact of image processing parameters was then explored using a CHO with 10 Laguerre-Gauss channels. In this work, we validate the CHO’s trend in nodule detectability as a function of two processing parameters by conducting a signal-known-exactly, multi-reader-multi-case (MRMC) ROC observer study. Five naive readers scored confidence of nodule visualization in 384 images with 50% nodule prevalence. The image backgrounds were regions-of-interest extracted from 6 normal patient scans, and the digitally inserted simulated nodules were obtained from phantom data in previous work. Each patient image was processed with both a near-optimal and a worst-case parameter combination, as determined by the CHO for nodule detection. The same 192 ROIs were used for each image processing method, with 32 randomly selected lung ROIs per patient image. Finally, the MRMC data was analyzed using the freely available iMRMC software of Gallas et al. Results: The image processing parameters which were optimized for the CHO led to a statistically significant improvement (p=0.049) in human observer AUC from 0.78 to 0.86, relative to the image processing implementation which produced the lowest CHO performance. Conclusion: Differences in user-selectable image processing methods on a commercially available digital radiography system were shown to have a marked impact on performance of human observers in the task of lung nodule detection. Further, the effect of processing on humans was similar to the effect on CHO performance. Future work will expand this study to include a wider range of detection/classification tasks and more observers, including experienced chest radiologists.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

TH-AB-209-01: Making Benchtop X-Ray Fluorescence Computed Tomography (XFCT) Practical for in Vivo Imaging by Integration of a Dedicated High-Performance X-Ray Source in Conjunction with Micro-CT Functionality

Purpose: To make benchtop x-ray fluorescence computed tomography (XFCT) practical for routine preclinical imaging tasks with gold nanoparticles (GNPs) by deploying, integrating, and characterizing a dedicated high-performance x-ray source and addition of simultaneous micro-CT functionality. Methods: Considerable research effort is currently under way to develop a polychromatic benchtop cone-beam XFCT system capable of imaging GNPs by stimulation and detection of gold K-shell x-ray fluorescence (XRF) photons. Recently, an ad hoc high-power x-ray source was incorporated and used to image the biodistribution of GNPs within a mouse, postmortem. In the current work, a dedicated x-ray source system featuring a liquid-cooled tungsten-target x-ray tube (max 160 kVp, ∼3 kW power) was deployed. The source was operated at 125 kVp, 24 mA. The tube’s compact dimensions allowed greater flexibility for optimizing both the irradiation and detection geometries. Incident x-rays were shaped by a conical collimator and filtered by 2 mm of tin. A compact “OEM” cadmium-telluride x-ray detector was implemented for detecting XRF/scatter spectra. Additionally, a flat panel detector was installed to allow simultaneous transmission CT imaging. The performance of the system was characterized by determining the detection limit (10-second acquisition time) for inserts filled with water/GNPs at various concentrations (0 and 0.010–1.0 wt%) and embedded in a small-animal-sized phantom. The phantom was loaded with 0.5, 0.3, and 0.1 wt% inserts and imaged using XFCT and simultaneous micro-CT. Results: An unprecedented detection limit of 0.030 wt% was experimentally demonstrated, with a 33% reduction in acquisition time. The reconstructed XFCT image accurately localized the imaging inserts. Micro-CT imaging did not provide enough contrast to distinguish imaging inserts from the phantom under the current conditions. Conclusion: The system is immediately capable of in vivo preclinical XFCT imaging with GNPs. Micro-CT imaging will require optimization of irradiation parameters to improve contrast. Supported by NIH/NCI grant R01CA155446; This investigation was supported by NIH/NCI grant R01CA155446.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

TH-AB-209-10: Breast Cancer Identification Through X-Ray Coherent Scatter Spectral Imaging

Purpose: We have previously described the development and testing of a coherent-scatter spectral imaging system for identification of cancer. Our prior evaluations were performed using either tissue surrogate phantoms or formalin-fixed tissue obtained from pathology. Here we present the first results from a scatter imaging study using fresh breast tumor tissues obtained through surgical excision. Methods: A coherent-scatter imaging system was built using a clinical X-ray tube, photon counting detectors, and custom-designed coded-apertures. System performance was characterized using calibration phantoms of biological materials. Fresh breast tumors were obtained from patients undergoing mastectomy and lumpectomy surgeries for breast cancer. Each specimen was vacuum-sealed, scanned using the scatter imaging system, and then sent to pathology for histological workup. Scatter images were generated separately for each tissue specimen and analyzed to identify voxels containing malignant tissue. The images were compared against histological analysis (H&E + pathologist identification of tumors) to assess the match between scatter-based and histological diagnosis. Results: In all specimens scanned, the scatter images showed the location of cancerous regions within the specimen. The detection and classification was performed through automated spectral matching without the need for manual intervention. The scatter spectra corresponding to cancer tissue were found to be in agreement with those reported in literature. Inter-patient variability was found to be within limits reported in literature. The scatter images showed agreement with pathologist-identified regions of cancer. Spatial resolution for this configuration of the scanner was determined to be 2–3 mm, and the total scan time for each specimen was under 15 minutes. Conclusion: This work demonstrates the utility of coherent scatter imaging in identifying cancer based on the scatter properties of the tissue. It presents the first results from coherent scatter imaging of fresh (unfixed) breast tissue using our coded-aperture scatter imaging approach for cancer identification.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2014

Quantitative Imaging with a Mobile Phone Microscope

Use of optical imaging for medical and scientific applications requires accurate quantification of features such as object size, color, and brightness. High pixel density cameras available on modern mobile phones have made photography simple and convenient for consumer applications; however, the camera hardware and software that enables this simplicity can present a barrier to accurate quantification of image data. This issue is exacerbated by automated settings, proprietary image processing algorithms, rapid phone evolution, and the diversity of manufacturers. If mobile phone cameras are to live up to their potential to increase access to healthcare in low-resource settings, limitations of mobile phone–based imaging must be fully understood and addressed with procedures that minimize their effects on image quantification. Here we focus on microscopic optical imaging using a custom mobile phone microscope that is compatible with phones from multiple manufacturers. We demonstrate that quantitative microscopy with micron-scale spatial resolution can be carried out with multiple phones and that image linearity, distortion, and color can be corrected as needed. Using all versions of the iPhone and a selection of Android phones released between 2007 and 2012, we show that phones with greater than 5 MP are capable of nearly diffraction-limited resolution over a broad range of magnifications, including those relevant for single cell imaging. We find that automatic focus, exposure, and color gain standard on mobile phones can degrade image resolution and reduce accuracy of color capture if uncorrected, and we devise procedures to avoid these barriers to quantitative imaging. By accommodating the differences between mobile phone cameras and the scientific cameras, mobile phone microscopes can be reliably used to increase access to quantitative imaging for a variety of medical and scientific applications.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2020

RockFlow: Fast Generation of Synthetic Source Rock Images Using Generative Flow Models

Image-based evaluation methods are a valuable tool for source rock characterization. The time and resources needed to obtain images has spurred development of machine-learning generative models to create synthetic images of pore structure and rock fabric from limited image data. While generative models have shown success, existing methods for generating 3D volumes from 2D training images are restricted to binary images and grayscale volume generation requires 3D training data. Shale characterization relies on 2D imaging techniques such as scanning electron microscopy (SEM), and grayscale values carry important information about porosity, kerogen content, and mineral composition of the shale. Here, we introduce RockFlow, a method based on generative flow models that creates grayscale volumes from 2D training data. We apply RockFlow to baseline binary micro-CT image volumes and compare performance to a previously proposed model. We also show the extension of our model to 2D grayscale data by generating grayscale image volumes from 2D SEM and dual modality nanoscale shale images. The results show that our method underestimates the porosity and surface area on the binary baseline datasets but is able to generate realistic grayscale image volumes for shales. With improved binary data preprocessing, we believe that our model is capable of generating synthetic porous media volumes for a very broad class of rocks from shale to carbonates to sandstone.

58 GEOSCIENCES↗

DOE OSTI2019

First M87 Event Horizon Telescope Results. IV. Imaging the Central Supermassive Black Hole

We present the first Event Horizon Telescope (EHT) images of M87, using observations from April 2017 at 1.3 mm wavelength. These images show a prominent ring with a diameter of ~40 μas, consistent with the size and shape of the lensed photon orbit encircling the "shadow" of a supermassive black hole. The ring is persistent across four observing nights and shows enhanced brightness in the south. To assess the reliability of these results, we implemented a two-stage imaging procedure. In the first stage, four teams, each blind to the others' work, produced images of M87 using both an established method (CLEAN) and a newer technique (regularized maximum likelihood). This stage allowed us to avoid shared human bias and to assess common features among independent reconstructions. In the second stage, we reconstructed synthetic data from a large survey of imaging parameters and then compared the results with the corresponding ground truth images. This stage allowed us to select parameters objectively to use when reconstructing images of M87. Across all tests in both stages, the ring diameter and asymmetry remained stable, insensitive to the choice of imaging technique. We describe the EHT imaging procedures, the primary image features in M87, and the dependence of these features on imaging assumptions.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS2005

Operation and Performance of the Mars Exploration Rover Imaging System on the Martian Surface

This slide presentation details the Mars Exploration Rover (MER) imaging system. Over 144,000 images have been gathered from all Mars Missions, with 83.5% of them being gathered by MER. Each Rover has 9 cameras (Navcam, front and rear Hazcam, Pancam, Microscopic Image, Descent Camera, Engineering Camera, Science Camera) and produces 1024 x 1024 (1 Megapixel) images in the same format. All onboard image processing code is implemented in flight software and includes extensive processing capabilities such as autoexposure, flat field correction, image orientation, thumbnail generation, subframing, and image compression. Ground image processing is done at the Jet Propulsion Laboratory's Multimission Image Processing Laboratory using Video Image Communication and Retrieval (VICAR) while stereo processing (left/right pairs) is provided for raw image, radiometric correction; solar energy maps,triangulation (Cartesian 3-spaces) and slope maps.

imaging↗

NASA NTRS2008

Dependence of Adaptive Cross-correlation Algorithm Performance on the Extended Scene Image Quality

Recently, we reported an adaptive cross-correlation (ACC) algorithm to estimate with high accuracy the shift as large as several pixels between two extended-scene sub-images captured by a Shack-Hartmann wavefront sensor. It determines the positions of all extended-scene image cells relative to a reference cell in the same frame using an FFT-based iterative image-shifting algorithm. It works with both point-source spot images as well as extended scene images. We have demonstrated previously based on some measured images that the ACC algorithm can determine image shifts with as high an accuracy as 0.01 pixel for shifts as large 3 pixels, and yield similar results for both point source spot images and extended scene images. The shift estimate accuracy of the ACC algorithm depends on illumination level, background, and scene content in addition to the amount of the shift between two image cells. In this paper we investigate how the performance of the ACC algorithm depends on the quality and the frequency content of extended scene images captured by a Shack-Hatmann camera. We also compare the performance of the ACC algorithm with those of several other approaches, and introduce a failsafe criterion for the ACC algorithm-based extended scene Shack-Hatmann sensors.

wave-front sensing and control↗

NASA NTRS2015

Retinal Image Quality Assessment for Spaceflight-Induced Vision Impairment Study

Long-term exposure to space microgravity poses significant risks for visual impairment. Evidence suggests such vision changes are linked to cephalad fluid shifts, prompting a need to directly quantify microgravity-induced retinal vascular changes. The quality of retinal images used for such vascular remodeling analysis, however, is dependent on imaging methodology. For our exploratory study, we hypothesized that retinal images captured using fluorescein imaging methodologies would be of higher quality in comparison to images captured without fluorescein. A semi-automated image quality assessment was developed using Vessel Generation Analysis (VESGEN) software and MATLAB® image analysis toolboxes. An analysis of ten images found that the fluorescein imaging modality provided a 36% increase in overall image quality (two-tailed p=0.089) in comparison to nonfluorescein imaging techniques.

retina↗

DOE OSTI2016

Investigation of Energy Selective Neutron Imaging at the PULSTAR Reactor

Neutron imaging is complementary to X-ray imaging where both are used to investigate the internal features of objects nondestructively. Historically, it only began three years after discovery of the neutron in 1932, however, neutron radiographs with reasonable quality were first made in the mid-fifties. Conventionally, the image is produced by placing the object between the neutron source and a two-dimensional position sensitive detector. The source commonly has a full energy spectrum (white beam) of thermal or cold neutrons to obtain high intensity in the sample and short exposure time. As a result, all neutron energies contribute to the obtained radiograph. Since the interaction of the neutron with matter is energy dependent, different energies produce different contrast on the radiographic image, hence, additional information about the object can be obtained. Neutron imaging that utilizes a narrow spectrum of energy instead of white beam is called energy selective neutron imaging. Many techniques have been explored to select a narrow band of neutron energy for the use in imaging such as the use velocity selectors, double crystal Monochromators, and time of flight techniques (at pulsed neutron sources). The PULSTAR reactor of North Carolina State University (1 MWth) currently has an operational neutron powder diffractometer (NPD) installed on beam tube no. 4, which was designed for neutron scattering experiments and can provide a monoenergetic beam of neutrons. The neutron energy could be selected by diffraction means using a Si bent perfect crystal as a Monochromator. In this work, an investigation of energy selective neutron imaging at the NPD is discussed in terms of the system performance and the ability of material discrimination on a test sample by observing the contrast variations of the neutron radiographs obtained using different neutron energies. Energy selective neutron imaging experiment was successfully tested at the PULSTAR reactor's NPD. Although, the uniform region is small and the exposure time is long in comparison with the NIF, both facilities exhibit similar spatial resolution which is a very important characteristic for any imaging facility. The contrast variation of neuron radiographs for the test sample obtained at different energies indicated a potential advantage for using energy selective neutron imaging over the white beam. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

DOE OSTI2022

Noninvasive Imaging of a Liquid Jet

Liquid jets are found in many applications, from printing to manufacturing to entertainment. Herein this study uses three different noninvasive imaging modalities to compare resulting images of a liquid jet operating at three Reynolds numbers that cover laminar, transitional, and turbulent flow. Selected measurement quantities from each image type are also compared. High-speed backlit (BL) imaging is a simple imaging technique found in many laboratories, and this is compared to two high-speed X-ray imaging techniques, white beam (WB) imaging and focused beam (FB) radiography. BL imaging can provide a wide field of view and is easy to implement, but it only shows the presence or absence of liquid. WB imaging can show detailed contours on the surface of the liquid jet, but the imaging region is much smaller. FB radiography produces a point-source measurement and can provide the quantitative, instantaneous local liquid path length, termed the equivalent path length (EPL). All three techniques provide similar measures of jet thickness, with the FB measurements having less variation. FB measurements can also provide detailed cross sections of the average liquid jet thickness at high spatial resolutions.

42 ENGINEERING↗

DOE OSTI2014

SU-E-I-72: First Experimental Study of On-Board CBCT Imaging Using 2.5MV Beam On a Radiotherapy Linac

Purpose: Varian TrueBeam version 2.0 comes with a new inline 2.5MV beam modality for image guided patient setup. In this work we develop an iterative volumetric image reconstruction technique specific to the beam and investigate the possibility of obtaining metal artifact free CBCT images using the new imaging modality. Methods: An iterative reconstruction algorithm with a sparse representation constraint based on dictionary learning is developed, in which both sparse projection and low dose rate (10 MU/min) are considered. Two CBCT experiments were conducted using the newly available 2.5MV beam on a Varian TrueBeam linac. First, a Rando anthropomorphic head phantom with and without a copper bar inserted in the center was scanned using both 2.5MV and kV (100kVp) beams. In a second experiment, an MRI phantom with many coils was scanned using 2.5MV, 6MV, and kV (100kVp) beams. Imaging dose and the resultant image quality is studied. Results: Qualitative assessment suggests that there were no visually detectable metal artifacts in MV CBCT images, compared with significant metal artifacts in kV CBCT images, especially in the MRI phantom. For a region near the metal object in the head phantom, the 2.5MV CBCT gave a more accurate quantification of the electron density compared with kV CBCT, with a ∼50% reduction in mean HU error. As expected, the contrast between bone and soft-tissue in 2.5MV CBCT decreased compared with kV CBCT. Conclusion: On-board CBCT imaging with the new 2.5MV beam can effectively reduce metal artifacts, although with a reduced softtissue contrast. Combination of kV and MV scanning may lead to metal artifact free CBCT images with uncompromised soft-tissue contrast.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

MO-FG-204-02: Reference Image Selection in the Presence of Multiple Scan Realizations

Purpose: Fusing information from multiple correlated realizations (e.g., 4DCT) can improve image quality. This process often involves ill-conditioned and asymmetric nonlinear registration and the proper selection of a reference image is important. This work proposes to examine post-registration variation indirectly for such selection, and develops further insights to reduce the number of cross-registrations needed. Methods: We consider each individual scan as a noisy point in the vicinity of an image manifold, related by motion. Nonrigid registration “transports” a scan along the manifold to the reference neighborhood, and the residual is a surrogate for local variation. To test this conjecture, 10 thoracic scans from the same session were reconstructed from a recently developed low-dose helical 4DCT protocol. Pairwise registration was repeated bi-directionally (81 times) and fusion was performed with each candidate reference. The fused image quality was assessed with SNR and CNR. Registration residuals in SSD, harmonic energy, and deformation Jacobian behavior were examined. The semi-symmetry is further utilized to reduce the number of registration needed. Results: The comparison of image quality between single image and fused ones identified reduction of local intensity variance as the major contributor of image quality, boosting SNR and CNR by 5 to 7 folds. This observation further suggests the criticality of good agreement across post-registration images. Triangle inequality on the SSD metric provides a proficient upper-bound and surrogate on such disagreement. Empirical observation also confirms that fused images with high residual SSD have lower SNR and CNR than the ones with low or intermediate SSDs. Registration SSD is structurally close enough to symmetry for reduced computation. Conclusion: Registration residual is shown to be a good predictor of post-fusion image quality and can be used to identify good reference centers. Semi-symmetry of the registration residual further reduces computation cost. Supported by in part by NIH R01 CA096679.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2023

Using Ultrasound Image Augmentation and Ensemble Predictions to Prevent Machine-Learning Model Overfitting

Deep learning predictive models have the potential to simplify and automate medical imaging diagnostics by lowering the skill threshold for image interpretation. However, this requires predictive models that are generalized to handle subject variability as seen clinically. Here, we highlight methods to improve test accuracy of an image classifier model for shrapnel identification using tissue phantom image sets. Using a previously developed image classifier neural network—termed ShrapML—blind test accuracy was less than 70% and was variable depending on the training/test data setup, as determined by a leave one subject out (LOSO) holdout methodology. Introduction of affine transformations for image augmentation or MixUp methodologies to generate additional training sets improved model performance and overall accuracy improved to 75%. Further improvements were made by aggregating predictions across five LOSO holdouts. This was done by bagging confidences or predictions from all LOSOs or the top-3 LOSO confidence models for each image prediction. Top-3 LOSO confidence bagging performed best, with test accuracy improved to greater than 85% accuracy for two different blind tissue phantoms. This was confirmed by gradient-weighted class activation mapping to highlight that the image classifier was tracking shrapnel in the image sets. Overall, data augmentation and ensemble prediction approaches were suitable for creating more generalized predictive models for ultrasound image analysis, a critical step for real-time diagnostic deployment.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2020

Super resolution for root imaging

Premise High‐resolution cameras are very helpful for plant phenotyping as their images enable tasks such as target vs. background discrimination and the measurement and analysis of fine above‐ground plant attributes. However, the acquisition of high‐resolution images of plant roots is more challenging than above‐ground data collection. An effective super‐resolution (SR) algorithm is therefore needed for overcoming the resolution limitations of sensors, reducing storage space requirements, and boosting the performance of subsequent analyses. Methods We propose an SR framework for enhancing images of plant roots using convolutional neural networks. We compare three alternatives for training the SR model: (i) training with non‐plant‐root images, (ii) training with plant‐root images, and (iii) pretraining the model with non‐plant‐root images and fine‐tuning with plant‐root images. The architectures of the SR models were based on two state‐of‐the‐art deep learning approaches: a fast SR convolutional neural network and an SR generative adversarial network. Results In our experiments, we observed that the SR models improved the quality of low‐resolution images of plant roots in an unseen data set in terms of the signal‐to‐noise ratio. We used a collection of publicly available data sets to demonstrate that the SR models outperform the basic bicubic interpolation, even when trained with non‐root data sets. Discussion The incorporation of a deep learning–based SR model in the imaging process enhances the quality of low‐resolution images of plant roots. We demonstrate that SR preprocessing boosts the performance of a machine learning system trained to separate plant roots from their background. Our segmentation experiments also show that high performance on this task can be achieved independently of the signal‐to‐noise ratio. We therefore conclude that the quality of the image enhancement depends on the desired application.

Ruiz‐Munoz, Jose F.↗

DOE OSTI2017

Investigating the surface dose contribution of intrafractional kV imaging in CyberKnife-based stereotactic radiosurgery

CyberKnife treatment consists of hundreds of noncoplanar beams and numerous intrafractional images that can be taken during a single treatment fraction; thus, doses because of imaging should be considered in this technique. The aim of this study is to investigate the in-field and out-of-field surface doses induced from kV imaging system during stereotactic radiosurgery (SRS) treatment. The imaging-induced surface doses were measured at the center of the imaging field and within ±15-cm distance from the center in both craniocaudal and lateral directions. TLD100H thermoluminescence dosimeters and EBT2 gafchromic films were used to take the measurements at the locations of 0, ±5, ±10, and ±15 cm in the 2 orthogonal directions on abdominal region of a Rando phantom. The surface dose contributions of imaging system for the 4 most commonly used energy options of 90, 100, 110, and 120 kVp with 3 mAs options of 10, 30, and 90 mAs were measured and compared. Imaging dose values have a positive correlation with both parameters of energy and mAs. The energy options of 100, 110, and 120 kVp, in average, induced 60%, 101%, and 141% more doses per mAs than 90 kVp energy in the imaging field center. A threefold increase in mAs values, i.e., from 10 mAs to 30 mAs and from 30 mAs to 90 mAs, caused higher dose in field center with a factor of 2.53 ± 0.08 when the energy value was kept constant. The in-field dose distributions within ±10 cm in both directions showed a flat pattern with a standard deviation lower than 5%, whereas the out-of-field doses at ±15-cm distance from the field center suddenly dropped to almost half of the central doses. Although a single imaging attempt causes a very low dose compared with the therapeutic dose level, one should be aware of the cumulative surface dose increase with higher imaging number. Proper patient setup, fiducial usage, and reduction of both the mAs values and the imaging numbers should be, therefore, considered to keep the cumulative surface dose in a lower level.

61 RADIATION PROTECTION AND DOSIMETRY↗

DOE OSTI2021

2D-to-3D image translation of complex nanoporous volumes using generative networks

Image-based characterization offers a powerful approach to studying geological porous media at the nanoscale and images are critical to understanding reactive transport mechanisms in reservoirs relevant to energy and sustainability technologies such as carbon sequestration, subsurface hydrogen storage, and natural gas recovery. Nanoimaging presents a trade off, however, between higher-contrast sample-destructive and lower-contrast sample-preserving imaging modalities. Furthermore, high-contrast imaging modalities often acquire only 2D images, while 3D volumes are needed to characterize fully a source rock sample. In this work, we present deep learning image translation models to predict high-contrast focused ion beam-scanning electron microscopy (FIB-SEM) image volumes from transmission X-ray microscopy (TXM) images when only 2D paired training data is available. We introduce a regularization method for improving 3D volume generation from 2D-to-2D deep learning image models and apply this approach to translate 3D TXM volumes to FIB-SEM fidelity. We then segment a predicted FIB-SEM volume into a flow simulation domain and calculate the sample apparent permeability using a lattice Boltzmann method (LBM) technique. Results show that our image translation approach produces simulation domains suitable for flow visualization and allows for accurate characterization of petrophysical properties from non-destructive imaging data.

58 GEOSCIENCES↗

DOE OSTI2026

INR-TEM: Robust cavity detection in multifocus TEM images via implicit neural representations

When characterizing materials using transmission electron microscopy (TEM) images, detecting and quantifying small features in microstructures, such as cavities, pose significant challenges. Off-the-shelf object detection models, including YOLOv8, show considerable performance degradation, particularly when images vary in resolution and the objects of interest possess a low percentage of the total image region of interest. In this study, we introduce a novel detection pipeline that incorporates an implicit neural representation (INR)-based detection method, INR-TEM, and two-modality imaging (e.g., under-focused and over-focused images typically acquired during materials characterization) to improve object detection performance. The INR-TEM method incorporates a pixel-wise prediction principle inspired by pixel-wise centerness weighting. INR-TEM demonstrates superior robustness to resolution variability, maintaining high detection accuracy even at low image resolutions compared to YOLOv8. To leverage INR-TEM effectively in real-world two-modality characterization applications, we further integrate a two-stage motion correction pipeline designed explicitly for aligning multifocus TEM images. The alignment process, comprising keypoint (based on scale-invariant feature transform, SIFT) and intensity matching, significantly mitigates the adverse effects of perceived motion-induced image degradation during through-focal TEM imaging, directly enhancing INR-TEM’s detection capability over conventional single-focus images. Our integrated INR-TEM cavity detection framework notably improves performance across various cavity sizes, outperforming off-the-shelf YOLOv8 detections that rely on a single image modality.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗