Search NASA⌕ Search

SEARCH · Search NASA

Results for “small target”

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.

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

At least 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Deep-learning-based workflow for boundary and small target segmentation in digital rock images using UNet++ and IK-EBM

We report three-dimensional (3D) X-ray micro-computed tomography (μCT) has been widely used in petroleum engineering because it can provide detailed pore structural information for a reservoir rock, which can be imported into a pore-scale numerical model to simulate the transport and distribution of multiple fluids in the pore space. The partial volume blurring (PVB) problem is a major challenge in segmenting raw μCT images of rock samples, which impacts boundaries and small targets near the resolution limit. We developed a deep-learning (DL)-based workflow for accurate and fast partial volume segmentation. The DL model's performance depends primarily on the training data quality and model architecture. This study employed the entropy-based-masking indicator kriging (IK-EBM) to segment 3D Berea sandstone images as training datasets. The comparison between IK-EBM and manual segmentation using a 3D synthetic sphere pack, which had a known ground truth, showed that IK-EBM had higher accuracy on partial volume segmentation. We then trained and tested the UNet++ model, a state-of-the-art supervised encoder-decoder model, for binary (i.e., void and solid) and four-class segmentation. We compared the UNet++ with the commonly used U-Net and wide U-Net models and showed that the UNet++ had the best performance in terms of pixel-wise and physics-based evaluation metrics. Specifically, boundary-scaled accuracy demonstrated that the UNet++ architecture outperformed the regular U-Net architecture in the segmentation of pixels near boundaries and small targets, which were subjected to the PVB effect. Feature map visualization illustrated that the UNet++ bridged the semantic gaps between the feature maps extracted at different depths of the network, thereby enabling faster convergence and more accurate extraction of fine-scale features. The developed workflow significantly enhances the performance of supervised encoder-decoder models in partial volume segmentation, which has extensive applications in fundamental studies of subsurface energy, water, and environmental systems.

02 PETROLEUM↗
DOE OSTI2020

Genomic discovery of an evolutionarily programmed modality for small-molecule targeting of an intractable protein surface

The vast majority of intracellular protein targets are refractory toward small-molecule therapeutic engagement, and additional therapeutic modalities are needed to overcome this deficiency. Here, the identification and characterization of a natural product, WDB002, reveals a therapeutic modality that dramatically expands the currently accepted limits of druggability. WDB002, in complex with the FK506-binding protein (FKBP12), potently and selectively binds the human centrosomal protein 250 (CEP250), resulting in disruption of CEP250 function in cells. The recognition mode is unprecedented in that the targeted domain of CEP250 is a coiled coil and is topologically featureless, embodying both a structural motif and surface topology previously considered on the extreme limits of “undruggability” for an intracellular target. Structural studies reveal extensive protein–WDB002 and protein–protein contacts, with the latter being distinct from those seen in FKBP12 ternary complexes formed by FK506 and rapamycin. Outward-facing structural changes in a bound small molecule can thus reprogram FKBP12 to engage diverse, otherwise “undruggable” targets. The flat-targeting modality demonstrated here has the potential to expand the druggable target range of small-molecule therapeutics. As CEP250 was recently found to be an interaction partner with the Nsp13 protein of the SARS-CoV-2 virus that causes COVID-19 disease, it is possible that WDB002 or an analog may exert useful antiviral activity through its ability to form high-affinity ternary complexes containing CEP250 and FKBP12.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Automatic Classification of Biological Targets in a Tidal Channel Using a Multibeam Sonar

Multibeam sonars are widely used for environmental monitoring of fauna at marine renewable energy sites. However, they can rapidly accrue vast volumes of data, which poses a challenge for data processing. Here, using data from a deployment in a tidal channel with peak currents of 1–2 m s –1 , we demonstrate the data-reduction benefits of real-time automatic classification of targets detected and tracked in multibeam sonar data. First, we evaluate classification capabilities for three machine learning algorithms: random forests, support vector machines, and k-nearest neighbors. For each algorithm, a hill-climbing search optimizes a set of hand-engineered attributes that describe tracked targets. Here, the random forest algorithm is found to be most effective—in postprocessing, discriminating between biological and nonbiological targets with a recall rate of 0.97 and a precision of 0.60. In addition, 89% of biological targets are correctly classified as either seals, diving birds, fish schools, or small targets. Model dependence on the volume of training data is evaluated. Second, a real-time implementation of the model is shown to distinguish between biological targets and nonbiological targets with nearly the same performance as in postprocessing. From this, we make general recommendations for implementing real-time classification of biological targets in multibeam sonar data and the transferability of trained models.

16 TIDAL AND WAVE POWER↗
DOE OSTI2022

A Topological Approach for Motion Track Discrimination

Detecting small targets at range is difficult because there is not enough spatial information present in an image sub-region containing the target to use correlation-based methods to differentiate it from dynamic confusers present in the scene. Moreover, this lack of spatial information also disqualifies the use of most state-of-the-art deep learning image-based classifiers. Here, we use characteristics of target tracks extracted from video sequences as data from which to derive distinguishing topological features that help robustly differentiate targets of interest from confusers. In particular, we calculate persistent homology from time-delayed embeddings of dynamic statistics calculated from motion tracks extracted from a wide field-of-view video stream. In short, we use topological methods to extract features related to target motion dynamics that are useful for classification and disambiguation and show that small targets can be detected at range with high probability.

Emerson, Tegan H.↗
DOE OSTI2023

Gamma-ray Measurements from Neutron Pulsed Die-Away Experiments

Pulsed-neutron die-away experiments are a promising experimental method that complement criticality benchmarks for nuclear data validation. The experiments have several advantages that include compatibility with non-fissile material, low benchmark uncertainties, and high sensitivity to absorption and scattering cross sections. When validating thermal neutron scattering laws, small targets are desirable for their large sensitivity to the scattering cross section and angular distribution. Unfortunately, room return or scattering from a shielding box limits the use of the very small targets needed to maximize sensitivity to thermal scattering laws. To address this problem, we propose changing the observable of the experiment from leaked thermal neutrons to γ-rays produced by (n, γ) reactions in the target. If feasible, this change allows removal of the shielding box, achieving higher sensitivity to thermal neutron scattering laws. This study focuses on replicating benchmark pulsed-neutron die-away experiments with γ-rays. We compare the integral parameters with γ-rays to those obtained with neutrons. We also discuss experimental design choices such as detector placement, shielding, and the presence of Cd. The results show the integral parameter with γ-rays is sensitive to detector location with respect to the target when Cd shielding is present. In addition, while Pb shielding around the γ-ray detectors do not seem to affect the integral parameter, the presence of Cd shielding does and adds background γ-rays to the die-away curves making it difficult to calculate the integral parameter.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

Automatic Detection and Classification of Radio Galaxy Images by Deep Learning

Abstract Surveys conducted by radio astronomy observatories, such as SKA, MeerKAT, Very Large Array, and ASKAP, have generated massive astronomical images containing radio galaxies (RGs). This generation of massive RG images has imposed strict requirements on the detection and classification of RGs and makes manual classification and detection increasingly difficult, even impossible. Rapid classification and detection of images of different types of RGs help astronomers make full use of the observed astronomical image data for further processing and analysis. The classification of FRI and FRII is relatively easy, and there are more studies and literature on them at present, but FR0 and FRI are similar, so it is difficult to distinguish them. It poses a greater challenge to image processing. At present, deep learning has made breakthrough progress in the field of image analysis and processing and has preliminary applications in astronomical data processing. Compared with classification algorithms that can only classify galaxies, object detection algorithms that can locate and classify RGs simultaneously are preferred. In target detection algorithms, YOLOv5 has outstanding advantages in the classification and positioning of small targets. Therefore, we propose a deep-learning method based on an improved YOLOv5 object detection model that makes full use of multisource data, combining FIRST radio with SDSS optical image data, and realizes the automatic detection of FR0, FRI, and FRII RGs. The innovation of our work is that on the basis of the original YOLOv5 object detection model, we introduce the SE Net attention mechanism, increase the number of preset anchors, adjust the network structure of the feature pyramid, and modify the network structure, thereby allowing our model to demonstrate galaxy classification and position detection effects. Our improved model produces satisfactory results, as evidenced by experiments. Overall, the mean average precision (mAP@0.5) of our improved model on the test set reaches 89.4%, which can determine the position (R.A. and decl.) and automatically detect and classify FR0s, FRIs, and FRIIs. Our work contributes to astronomy because it allows astronomers to locate FR0, FRI, and FRII galaxies in a relatively short time and can be further combined with other astronomically generated data to study the properties of these galaxies. The target detection model can also help astronomers find FR0s, FRIs, and FRIIs in future surveys and build a large-scale star RG catalog. Moreover, our work is also useful for the detection of other types of galaxies.

Astronomy & Astrophysics↗
DOE OSTI2020

Northstar Mo100 Target Beam Spot Size Sensitivity Study

Northstar Mo100 target for production of the medical isotope Mo99 is particularly sensitive to beam over focusing because of the intended high power density. At 38 MeV and 6.32 mA (240 kW beam power, nominally 160 kW th in the target), small deviation from the nominal 12 mm FWHM beam can significantly increase the target disk temperatures and more importantly the load bearing target window temperature. A rhodotron will be the source of the electron beam to be used in the isotope production. This machine operates in a pulsed beam mode, currently expected to be at 50 Hz, which results in a 100 C temperature oscillation above and below the steady state temperature that is the focus of this study. This must be considered in the determination of minimum beam spot size that can be accepted by the target without risk of failure. In addition to the nominal beam spot size of 12 mm, the analysis of 10, 11 and 14.5 mm beam spot sizes are analyzed. The primary conclusion is that an over focused beam to just 11 mm FWHM is the minimum beam spot allowable, based on window temperature and the Ultimate Tensile Stress (UTS) of the window material Inconel 718 as a function of temperature. In contrast an under focused beam of 14.5 mm FWHM has much reduced target and window temperatures, but there is a resulting 6% drop in isotope production, all other factor staying the same. These results are presented herein.

07 ISOTOPE AND RADIATION SOURCES↗
DOE OSTI2026

Improvements in Mirror Surface Measurement with Reflected Computer Vision Targets

Over the last several years, NREL has been developing a system to measure large optical surfaces of heliostat mirrors by reflecting computer vision targets. An advantage of this system, called Reflected Target Nonintrusive Assessment (ReTNA), is that it lends itself well to stitching together many images, each reflecting only part of a larger heliostat. In the last few months, this was taken to a new extreme, with a small target (<5m2) being used to measure a >25m2 long focal length heliostat. These measurements were compared with traditional fringe deflectometry methods, which require a >50m2 target, and photogrammetry. The strengths, weaknesses and limitations of ReTNA are discussed. An estimated uncertainty in this new measurement is presented, along with software improvements and a new wireless data collection system. A bill of materials for this measurement system is presented, which has been designed to use all low-cost, off-the-shelf components. Finally, the next steps for future ReTNA development are presented. Overall, ReTNA can be a valuable optics measurement system, complimentary to existing measurement techniques available for large reflective surfaces.

14 SOLAR ENERGY↗
DOE OSTI2024

Comparative Assessment of U-Net-Based Deep Learning Models for Segmenting Microfractures and Pore Spaces in Digital Rocks

Segmentation of high-resolution X-ray microcomputed tomography (µCT) images is crucial in digital rock physics (DRP), affecting the characterization and analysis of microscale phenomena in the porous media. The complexity of geological structures and nonideal scanning conditions pose significant challenges to conventional image segmentation approaches. Motivated by the recent increasing popularity of deep learning (DL) techniques in image processing, this work undertakes a comparative study of DL models, specifically U-Net and its variants, for segmenting multiple targets with distinguished features in digital rocks, including discrete fracture networks (DFNs), pore spaces, and solid rock. Particularly, DFNs have a smaller volumetric fraction over others, bringing in a substantial challenge of imbalanced segmentation. The primary focus is to evaluate the architecture and feature enhancement strategies of various DL models, including U-Net, attention U-Net, residual U-Net, U-Net++, and residual U-Net++. The models were designed as 2.5D, utilizing a central 2D image and its two adjacent upper and lower 2D images as input to provide a pseudo-3D context. In addition, because the ground truth of segmentation was unknown for real-world digital rocks, we created a benchmark data set following the inverse operations of segmentation. The data synthesis started from the label images (i.e., solid rock, pore spaces, and DFNs), followed by simulating partial volume blurring, adding random background noise, and introducing ring artifacts to mimic real raw X-ray µCT images. The data set, which included various rock types (i.e., sandstone and artificial data), scanning resolution, and magnitudes of noise and artifacts, was divided into training and testing data sets with a 90% and 10% ratio, respectively. Moreover, in addition to the conventional pixel-wise evaluation metrics, the physics-based metric of the lattice-Boltzmann method (LBM) simulated permeability provided more comprehensive assessments. The results demonstrated that the residual connections, nested architectures, and redesigned skip connections contribute to the model performance and give the residual U-Net++ the highest accuracy. The improvements were mainly on the boundaries and small targets, especially the DFNs, which dominate the interconnectivity and therefore affect the permeability greatly. This study also rigorously evaluated the efficiency and generalization of each model, demonstrating that the sophisticated architectures achieved excellent practicability and maintained robust performance on completely unseen data, ensuring their suitability for diverse and challenging DRP applications.

58 GEOSCIENCES↗
DOE OSTI2025

Robustness of Deep Learning Classification to Adversarial Input on GPUs: Asynchronous Parallel Accumulation Is a Source of Vulnerability

The ability of machine learning (ML) classification models to resist small, targeted input perturbations—known as adversarial attacks—is a key measure of their safety and reliability. We show that floating-point non associativity (FPNA) coupled with asynchronous parallel programming on GPUs is sufficient to result in misclassification, without any perturbation to the input. Additionally, we show that this misclassification is particularly significant for inputs close to the decision boundary and that standard adversarial robustness results may be overestimated up to 4.6 when not considering machine-level details. We first study a linear classifier, before focusing on standard Graph Neural Network (GNN) architectures and datasets used in robustness assessments. We develop a novel black-box attack using Bayesian optimization to discover external workloads that can change the instruction scheduling which bias the output of reductions on GPUs and reliably lead to misclassification. Motivated by these results, we present a new learnable permutation (LP) gradient-based approach to learning floating-point operation orderings that lead to misclassifications. The LP approach provides a worst-case estimate in a computationally efficient manner, avoiding the need to run identical experiments tens of thousands of times over a potentially large set of possible GPU states or architectures. Finally, using instrumentation-based testing, we investigate parallel reduction ordering across different GPU architectures under external background workloads, when utilizing multi-GPU virtualization, and when applying power capping. Our results demonstrate that parallel reduction ordering varies significantly across architectures under the first two conditions, substantially increasing the search space required to fully test the effects of this parallel scheduler-based vulnerability. These results and the methods developed here can help to include machine-level considerations into adversarial robustness assessments, which can make a difference in safety and mission critical applications.

Shanmugavelu, Sanjif [Maxeler Technologies, a Groq↗
DOE OSTI2024

Demonstration of a hydrodynamically equivalent burning plasma in direct-drive inertial confinement fusion

Focussing laser light onto the surface of a small target filled with deuterium and tritium implodes it and leads to the creation of a hot and dense plasma, in which thermonuclear fusion reactions occur. In order for the plasma to become self-sustaining, the heating of the plasma must be dominated by the energy provided by the fusion reactions—a condition known as a burning plasma. A metric for this is the generalized Lawson parameter, where values above around 0.8 imply a burning plasma. Here, we report on hydro-equivalent scaling of experimental results on the OMEGA laser system and show that these have achieved core conditions that reach a burning plasma when the central part of the plasma, the hotspot, is scaled in size by at least a factor of 3.9 ± 0.10, which would require a driver laser energy of at least 1.7 ± 0.13 MJ. In addition, we hydro-equivalently scale the results to the 2.15 MJ of laser energy available at the National Ignition Facility and find that these implosions reach 86% of the Lawson parameter required for ignition. Furthermore, our results support direct-drive inertial confinement fusion as a credible approach for achieving thermonuclear ignition and net energy in laser fusion.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Autonomous Grid Support Functionality in Variable-Speed Drive-Based End-Use Loads

High penetration of inverter-based resources into power systems causes lower inertia, and it becomes more challenging to maintain grid stability during contingency events. While conventional demand response and shedding at the load side suffer from slow response, impact by unavailability, and issues from communication, variable-speed drives (VSDs) are becoming strong candidates for fast frequency response (FFR) thanks to their fast response and reduced user impact from variable power adjustment capability. In this letter, we propose an autonomous frequency support control method for a residential refrigerator with a VSD, a promising target small end-use load for FFR. We analyze the effect of the proposed controller on grid stability, and we verify the performance through laboratory experiments with a commercially available refrigerator.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

Cryo-EM structure determination of small therapeutic protein targets at 3 Å-resolution using a rigid imaging scaffold

Cryoelectron microscopy (Cryo-EM) has enabled structural determination of proteins larger than about 50 kDa, including many intractable by any other method, but it has largely failed for smaller proteins. Here, we obtain structures of small proteins by binding them to a rigid molecular scaffold based on a designed protein cage, revealing atomic details at resolutions reaching 2.9 Å. We apply this system to the key cancer signaling protein KRAS (19 kDa in size), obtaining four structures of oncogenic mutational variants by cryo-EM. Importantly, a structure for the key G12C mutant bound to an inhibitor drug (AMG510) reveals significant conformational differences compared to prior data in the crystalline state. The findings highlight the promise of cryo-EM scaffolds for advancing the design of drug molecules against small therapeutic protein targets in cancer and other human diseases.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Discovery and Development of a Small-Molecule Inhibitor Targeting the GAS41 YEATS Domain in Nonsmall Cell Lung Cancer

Abstract GAS41 is frequently overexpressed in Non-Small Cell Lung Cancer (NSCLC). GAS41 contains a YEATS domain, which recognizes acetylated lysine residues on histones to recruit protein complexes and facilitate transcription. Suppression of GAS41 in NSCLC models inhibits cellular proliferation and markedly reduces tumor growth in mouse xenografts, justifying the development of small-molecule inhibitors. We have employed structure-based design and medicinal chemistry optimization to discover DLG-41, a submicromolar inhibitor binding to the GAS41 YEATS domain. DLG-41 potently disrupts the association of GAS41 YEATS with chromatin in mammalian cells and inhibits the proliferation of NSCLC cell lines with submicromolar potency without significantly affecting normal lung fibroblasts. DLG-41 induces more effective growth inhibition in A549 versus GAS41-knockout cells, demonstrating on-target activity. DLG-41 treatment upregulates the CDKN1A gene and downregulates pathways associated with lung cancer cell identity, tumor migration, and invasion. DLG-41 is a promising chemical probe for targeting GAS41 protein in NSCLC models and has potential for future development.

Listunov, Dymytrii [University of Michigan , , , ,↗