Improved StyleGAN-v2 based Inversion for Out-of-Distribution Images
Explore the source record for details and available documents.
Computer science↗
SEARCH · Engineering Papers
Results for “Computational 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.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Enforcing global constraints for the dispersion closure problem: τ 2 -SIMPLE algorithm
Permeability and effective dispersion tensors are critical parameters to characterize flow and transport in porous media at the continuum scale. Homogenization theory defines a framework in which such effective properties are first computed from solving a closure problem in a repeating unit cell of the periodic microstructure and then used in a macroscopic formulation for efficient computation. The closure problem is formulated as a local boundary value problem subjected to global constraints, which guarantee the uniqueness of the solution and can be difficult to satisfy for complex geometries and at high flow conditions. These constraints also ensure that pore-scale pressure, velocity, and concentration fields can be accurately reconstructed from the closure variable. Building on a previous work, here we present a framework that allows to satisfy global constraints associated to both the permeability and the dispersion closure problems by introducing two artificial time scales. The algorithm, called τ 2 -SIMPLE, computes both permeability and effective dispersion given an arbitrarily complex geometry and flow condition. Furthermore, this algorithm is demonstrated to be accurate for both 2D and 3D geometries across varying flow conditions, and thus it can be used to quickly characterize effective properties from porous media images in many applications.
97 MATHEMATICS AND COMPUTING↗
Toward memory-efficient melt pool monitoring: a classification framework using event-based imaging and sparse sensing technique
Vision sensors like CMOS and CCD cameras are often used for in-process monitoring of melt pools in laser-based additive and welding processes, but they require transferring large amounts of data and computational processing resources. Event-based neuromorphic imagery, on the other hand, detects only the change in pixel intensity, thus potentially reducing the data amount and latency. With an event imager, this study develops a framework for melt pool condition classification, including image construction, time scale selection, optimal pixel selection, and sparse classification, to achieve a highly memory-efficient scheme. These are based on sparse sensing techniques with singular value decomposition (SVD) and QR pivoting, the two fundamental matrix transformations for linear dimensionality reduction. The framework is then validated by classifying a controlled experiment by exciting various mode shapes of liquid gallium pools of varying depths (3, 6, and 8 mm). At 200 pixels, the classifier can reach overall accuracy of 75%, while at 2000 pixels (0.013% of the total possible pixels), the accuracy is nearly 90% (89.86%). At the same number of pixels, random selection can only achieve 46% and 67%, respectively. The memory savings of the sparsely sampled event data compared to a conventional imager is about 500 times. In addition to performance, implementation and limitations of the framework are also discussed.
42 ENGINEERING↗
ODIN: Probing the LAE Ly α Luminosity Function across Cosmic Time and Different Environments
The ubiquity and relative ease of discovery make 2 ≲ z ≲ 5 Lyα emitting galaxies (LAEs) ideal tracers for large-scale structure of the distant Universe. In addition, because Lyα is a resonance line, but frequently observed at large equivalent width, it is potentially a probe of galaxy evolution. The LAE Lyα luminosity function (LF) is an essential measurement for making progress on both of these topics. Although several studies have computed the LAE LF, very few have delved into how the function varies with environment. The large area and depth of the One-hundred-deg 2 DECam Imaging in Narrowbands (ODIN) survey makes such measurements possible at the cosmic noon redshifts of z ∼ 2.4, 3.1, and 4.5. In this initial work, we present algorithms needed to rigorously compute the LAE LF, and test them on the ∼16,000 ODIN LAEs found in the extended COSMOS field. Using these limited samples, we find weak evidence that protocluster environments suppress the numbers of faint LAEs compared to the field. We also find that the LF decreases in number density and evolves towards a steeper faint-end slope over cosmic time from z ∼ 4.5 to z ∼ 2.4.
Lyman-alpha galaxies↗
AI/ML Expo Boosting Job Performance with AI: Innovative Approaches and Success Stories
Our technology leverages artificial intelligence (AI) to enhance the user experience in High Performance Computing (HPC) environments. By analyzing user behavior and providing personalized recommendations, our AI system helps HPC users optimize their workflows and improve productivity. Additionally, we offer an advanced image similarity search feature, which utilizes AI algorithms to identify and retrieve visually similar images, saving users valuable time and effort in their research and analysis.
97 - MATHEMATICS AND COMPUTING↗
Sparse Matrix-Based HPC Tomography
Tomographic imaging has benefited from advances in X-ray sources, detectors and optics to enable novel observations in science, engineering and medicine. These advances have come with a dramatic increase of input data in the form of faster frame rates, larger fields of view or higher resolution, so high performance solutions are currently widely used for analysis. Tomographic instruments can vary significantly from one to another, including the hardware employed for reconstruction: from single CPU workstations to large scale hybrid CPU/GPU supercomputers. Furthermore, flexibility on the software interfaces and reconstruction engines are also highly valued to allow for easy development and prototyping. This paper presents a novel software framework for tomographic analysis that tackles all aforementioned requirements. The proposed solution capitalizes on the increased performance of sparse matrix-vector multiplication and exploits multi-CPU and GPU reconstruction over MPI. Furthermore, the solution is implemented in Python and relies on CuPy for fast GPU operators and CUDA kernel integration, and on SciPy for CPU sparse matrix computation. As opposed to previous tomography solutions that are tailor-made for specific use cases or hardware, the proposed software is designed to provide flexible, portable and high-performance operators that can be used for continuous integration at different production environments, but also for prototyping new experimental settings or for algorithmic development. The experimental results demonstrate how our implementation can even outperform state-of-the-art software packages used at advanced X-ray sources worldwide.
97 MATHEMATICS AND COMPUTING↗
Frequency-domain computing using nonlinear acoustic-wave device on lithium niobate
Abstract Multiply-accumulation are crucial computing operations in signal processing, numerical simulations, and machine learning. In recent years, optical analog approaches have demonstrated higher computing performance and better power efficiency than their digital counterparts. However, analog computing chips usually need large areas and complex structures for parallel computing, as a single device element only executes one computing operation at a single time. Here, we demonstrate frequency-domain computing using the nonlinear acoustic-wave devices on lithium niobate, featuring a normalized external second-harmonic generation conversion efficiency of ~ 5.7 × 10-4 W-1. The second-order sum-frequency nonlinear process of lithium niobate enables multiplication of inputs encoded in the frequency domain. Compared to the analog schemes, our device features a notably simpler design, and nanofabrication requires only one lift-off. Using a single acoustic-wave device within an area of 0.03 mm2, we can simultaneously conduct over 130,000 multiply-accumulation operations. Our acoustic-wave device shows applications in real and complex vector convolutions and image processing. This demonstration sets the stage for experimental realizations into frequency-domain integrated nonlinear acoustic computing systems, potentially shaping future developments in acoustic neural networks and quantum computing.
chai, mingzhao (ORCID:0009000466226341)↗
Cross-correlation image analysis for real-time single particle tracking
Accurately measuring the translations of objects between images is essential in many fields, including biology, medicine, chemistry, and physics. One important application is tracking one or more particles by measuring their apparent displacements in a series of images. Popular methods, such as the center of mass, often require idealized scenarios to reach the shot noise limit of particle tracking and, therefore, are not generally applicable to multiple image types. More general methods, such as maximum likelihood estimation, reliably approach the shot noise limit, but are too computationally intense for use in real-time applications. These limitations are significant, as real-time, shot-noise-limited particle tracking is of paramount importance for feedback control systems. To fill this gap, we introduce a new cross-correlation-based algorithm that approaches shot-noise-limited displacement detection and a graphics processing unit-based implementation for real-time image analysis of a single particle.
Instruments & Instrumentation↗
Application of unsupervised deep learning to image segmentation and in-situ contact angle measurements in a CO 2 -water-rock system
Rock surface wettability is a critical property that regulates multiphase flows in porous media, which can be quantified using the surface contact angle (CA). X-ray micro-computed tomography (μCT) provides an effective approach to in-situ measurements of surface CAs. However, the CA measurement accuracy depends significantly on the quality of CT image segmentation, which is the clustering of CT pixels into separate phases. Inspired by this, we developed a deep learning (DL)-based CA measurement workflow. Motivated by the recent tremendous progress in unsupervised learning techniques and aiming to avoid expensive manual data annotations, an unsupervised DL pipeline for CT image segmentation was proposed and implemented, which includes unsupervised model training and post-processing. The unsupervised model training was driven by a novel loss function constrained with feature similarity and spatial continuity and implemented by iterative forward and backward paths; the former clustered the pixel-wise feature vectors extracted by convolution neural networks, whereas the latter updated the parameters using gradient descent. An over-segmentation strategy was adopted for model training. The post-processing steps based on agglomerative hierarchical clustering (AHC) were implemented to further merge the over-segmented model output to the desired cluster number, which is intended to improve the efficiency of image segmentation. The developed unsupervised DL pipeline was compared with other commonly-used image segmentation methods using pixel-wise and physics-based evaluation metrics on a synthetic raw-image dataset, which had a known ground truth. The unsupervised DL pipeline showed the best performance. Next, the segmented images were input to an automatic CA measurement tool, and the results were validated by comparisons with manual measurements. The CA values from the manual and automatic measurements showed similar distributions and statistical properties. The automatic measurement demonstrated a wider spectrum because of the much larger number of measurement data points. The primary novelty of the unsupervised DL pipeline developed in this study lies in the novel loss function and the over-segmentation strategy associated with AHC post-processing. Finally, the workflow has been proven an efficient tool for pore-scale wettability characterization, which has a wide range of applications in fundamental studies of multiphase flows in natural porous media, which have critical implications to geological carbon sequestration, hydrocarbon energy recovery, and contaminant transport in groundwater.
42 ENGINEERING↗
Direct observation of C 3 S particle dissolution using fast nano X-ray computed tomography
Tricalcium silicate (C 3 S) occupies 50 % to 70 % of ordinary portland cement (OPC) by mass and it is an important component affecting the hydration of OPC [1], [2], [3], [4], [5], [6], [7]. Generally, the hydration of C 3 S is described by two processes: the dissolution of C 3 S particles and the precipitation of hydration products. While it is understood that the dissolution rates of C 3 S vary with time, more precise measurements are needed to understand this process. Many mechanisms have been proposed to explain the time-evolving dissolution rates of C 3 S [2]. The metastable barrier hypothesis suggests that a thin metastable layer of hydrates forms around the C 3 S particle surface and prohibits the access of grains to the aqueous solution [8], [9], [10], [11], [12]. The slow dissolution step hypothesis suggests that the increased ion concentration from the initial reaction delays the C 3 S dissolution [2], [13], [14], [15], [16], [17]. More recent publications suggest that C 3 S may react differently depending on the existence of crystallographic defects [18], [19], [20]. Etch pits are thought to open on the particle surface during the initial reaction and this contributes to the C 3 S dissolution [21], [22]. As hydrates precipitate and cover these highly reactive surfaces, hydration slows down and the induction period starts [18], [23], [24], [25], [26]. Many experiments have been conducted to test the aforementioned mechanisms. Some hydration studies utilize bulk measurements, such as isothermal calorimetry [27], [28], [29], pore solution analysis [30], quasi-elastic neutron scattering [31], and nuclear magnetic resonance spectroscopy [32], [33]. One limitation of these measurements is that they do not provide direct and detailed information on the individual C 3 S particles. Some other studies utilize imaging techniques, such as scanning electron microscopy (SEM) [34], [35], [36] and transmission electron microscopy (TEM) [37]. However, SEM/TEM cannot track the evolution of individual particles throughout hydration [34], [35], [38], [39], [40] and they do not give insights into the microstructure of materials before hydration [34], [35], [39]. This makes it challenging to draw strong conclusions from only SEM or TEM observations. Synchrotron X-ray tomography techniques have been used more broadly in recent years to study cement hydration. They are not only non-destructive but also able to image a sample in full 3D with resolutions that can reach from micron to nanoscale. Nano computed tomography (nCT) is one technique that has been applied to study cement hydration at the nanoscale [26], [41]. A typical nCT can reach a pixel size from 15 to 65 nm, providing enough detail for observing features <1 μm. However, nCT often takes >0.5 h to finish one scan. This makes the application of this technique on continuous scans for in-situ observations challenging. Fast X-ray computerized tomography (fCT) is another technique that has shown success in studying the time-evolving cement microstructures [20], [42], [43], [44], [45], [46], [47], [48], [49]. Due to the high flux of the X-ray beam from the synchrotron ring, fCT allows a scan to be captured within 1 min at a pixel size of 1 μm. This allows a paste sample to be continuously scanned during the hydration process. However, the micron-sized resolutions limit does not provide detailed insights for particles <5 μm [20], [49]. Fortunately, the combination of nCT and fCT has allowed the development of fast nano X-ray computed tomography (fnCT). fnCT can capture a 3D data set in <2 min at a pixel size of 50 nm. This makes this procedure an exciting method to evaluate hydrating pastes. fnCT collects multiple X-ray radiographs at various rotation angles and generates a 3D model of the scanned sample, which is also referred to as a 3D tomography [50], [51]. In one tomography, the X-ray absorptions of different components (e.g., C 3 S and hydrates) differ as functions of density and chemistry [52], [53]. These X-ray absorption contrasts can be used to extract detailed information about the 3D microstructure [26], [54], [55]. In this paper, fnCT is used to collect time-lapse tomographs of hydrating C 3 S paste from 18 min after mixing to 7 h of hydration. The bulk measurements of anhydrous C 3 S, as well as the microstructural changes of individual C 3 S particles, are directly observed, quantified, and discussed. The dissolution behavior of C 3 S particles at various size scales is systematically analyzed and compared. This work aims to find the relationship between the size of C 3 S particle sizes and their dissolution rates. This provides significant insights into the early-age hydration of C 3 S on length and time scales not previously possible. Because of the magnitude of the data and the substantial amount of observations, this work will solely focus on the change in the anhydrous particles. Changes in the hydration products will be reported in future work.
42 ENGINEERING↗
Gradient-based surface nuclear magnetic resonance for groundwater investigation
In medical magnetic resonance imaging, spatial localization (imaging) is based upon the application of controlled magnetic field gradients on top of the main magnetic field to spatially modulate the frequency and/or phase of the nuclear magnetic resonance (NMR) signal across the volume of investigation. In this work, we have applied similar physical principles to produce controlled magnetic field gradients during surface NMR-based groundwater investigations. In this approach, a gradient pulse of variable amplitude or duration is applied immediately after the excitation pulse to cause predictable phase encoding of the NMR signal as a function of depth. This approach is also applicable to emerging surface NMR detection methods that use a prepolarization field with fast nonadiabatic turn-off to generate detectable NMR signals from the shallow subsurface. In this case, the gradient pulse is applied after terminating the prepolarization field and provides a heretofore unavailable means of localizing the NMR response as a function of depth. The application of gradients can also be combined with tip-angle-based modulation to yield higher imaging resolution than can be achieved through either gradient- or tip-angle-based imaging alone. We implemented this new gradient-based capability into a surface NMR gradient generation accessory that is compatible with the GMR-Flex instrument and developed surface NMR-specific forward modeling and linear inverse models. We validated the accuracy of this novel gradient-based sNMR technology using computer simulations, experiments using a small pool filled with a discrete layer of bulk water, and field experiments at well-characterized groundwater test sites along Ebey Island, WA, and Larned, KS. The gradient-based sNMR imaging observations were compared with high-resolution direct push NMR results observed at these sites. The results of computer simulations and field experiments indicate improvements in both the detection (signal-to-noise ratio) and spatial resolution of shallow subsurface water content using gradient-based surface NMR, compared with traditional surface NMR imaging methods.
Geochemistry & Geophysics↗
Raspberry Pi–powered temperature monitoring of growth chamber microclimates
While controlled environments are desirable for growing and measuring plants, growth chambers and greenhouses typically have microclimates that impact plant growth, development, and stress responses. Furthermore, opening and closing the doors of a controlled environment introduces variation in the environment, especially at temperature extremes, affecting both the measurements and the organisms within. Using multiple temperature data loggers to normalize results can be cost-prohibitive and rarely offers real-time feedback on temperature status. We used low-cost single-board computers, cameras, and temperature sensors to manage and capture growth chamber temperatures while acquiring plant image data. Detailed here are methods to document microclimates within a growth chamber so that data can be normalized to measured temperature information. This protocol describes a low-cost method for automated temperature monitoring, which enables both high-throughput measurements of temperature along with plant growth and stress responses via plant imaging.
Plant Sciences↗
Delamination-informed lifecycle decisions: A dielectric and machine learning framework for composite sorting and recycling
Composite materials are widely used in aerospace, marine, and automotive sectors due to their high strength-to-weight ratio and durability. However, their long-term reliability can be compromised by damage accumulation. Specifically, delamination initiation serves as a precursor to structural failure, which is often difficult to detect during damage inspection. Identifying and sorting delamination initiation in samples not only increases operational safety while providing critical information for end-of-life decisions, which influences both the service life extension value and the efficiency of fiber extraction during recycling. This research addresses two challenges: (1) developing a nondestructive, ex-situ framework to sort composite materials based on damage severity, particularly delamination, and (2) understanding how damage in composites influences resin removal during pyrolysis. Both experimental work and finite element analysis were performed to predict critical stress levels that are associated with delamination onset. Based on these results, three loading levels 50 %, 75 %, and 90 % of maximum stress, were selected for controlled experiments, generating composite samples with varying extents of damage for machine learning model training. Microscopic imaging of these samples confirmed the damage progression from matrix cracking to delamination, validating the computational predictions. We explored supervised machine learning using dielectric measurements to classify damage states. Preliminary results show an artificial neural network can identify early delamination which is a potential precursor to failure, with 94.44 % accuracy on our dataset. A parallel investigation into the effect of damage severity on pyrolysis recycling showed that heavily delaminated samples required significantly less energy for comparable matrix removal than undamaged samples.
dielectric variables↗
Deep learning for Alzheimer's disease: Mapping large-scale histological tau protein for neuroimaging biomarker validation
Abnormal tau inclusions are hallmarks of Alzheimer's disease and predictors of clinical decline. Several tau PET tracers are available for neurodegenerative disease research, opening avenues for molecular diagnosis in vivo. However, few have been approved for clinical use. Understanding the neurobiological basis of PET signal validation remains problematic because it requires a large-scale, voxel-to-voxel correlation between PET and (immuno) histological signals. Large dimensionality of whole human brains, tissue deformation impacting co-registration, and computing requirements to process terabytes of information preclude proper validation. We developed a computational pipeline to identify and segment particles of interest in billion-pixel digital pathology images to generate quantitative, 3D density maps. The proposed convolutional neural network for immunohistochemistry samples, IHCNet, is at the pipeline's core. We have successfully processed and immunostained over 500 slides from two whole human brains with three phospho-tau antibodies (AT100, AT8, and MC1), spanning several terabytes of images. Our artificial neural network estimated tau inclusion from brain images, which performs with ROC AUC of 0.87, 0.85, and 0.91 for AT100, AT8, and MC1, respectively. Introspection studies further assessed the ability of our trained model to learn tau-related features. We present an end-to-end pipeline to create terabytes-large 3D tau inclusion density maps co-registered to MRI as a means to facilitate validation of PET tracers.
60 APPLIED LIFE SCIENCES↗
Stress and Strain Heterogeneity and Persistence in Uniaxially‐ and Triaxially‐Loaded Sandstone
Two critical questions in brittle rock mechanics are how rocks developed localized strains and to what extent internal stress heterogeneity controls this localization and subsequent macroscopic failure. Definitive answers have not yet been found, but would provide insight into rock fracture mechanics as relevant to hydrocarbon extraction and sequestration. Here, we use synchrotron X‐ray tomography (XRT) and 3D X‐ray diffraction (3DXRD) during uniaxial and triaxial tests on Nugget and Bentheimer sandstones to examine strain and stress localization prior to mechanical failure. 3DXRD was used to measure intra‐granular lattice strains which were used to compute elastic stress tensors of each grain. Digital volume correlation (DVC) was applied to XRT images to determine the strain field in the sample. Both samples featured marked spatial heterogeneity, localization, and temporal persistence of elevated stresses and strains during their mechanical deformation toward failure. Both samples featured a majority of grains with at least one principal stress component that was tensile, a signature of the influence of heterogeneity on stress transmission. Measurements further revealed that compressive stress orientations and statistics evolved in a similar manner to those of inter‐particle forces in loose granular materials, with triaxially‐compressed rock exhibiting enhanced grain stress heterogeneity compared to uniaxially‐compressed rock. Our results complement recent work by others who employed XRT and scanning 3DXRD to study triaxially‐compressed sandstone, but extend those results to uniaxial compression, sandstones of varied porosity, and grain stress measurements throughout the 3D full extent of the samples rather than in a single layer examined with scanning 3DXRD.
58 GEOSCIENCES↗
Adaptive hyperparameter updating for training restricted Boltzmann machines on quantum annealers
Restricted Boltzmann Machines (RBMs) have been proposed for developing neural networks for a variety of unsupervised machine learning applications such as image recognition, drug discovery, and materials design. The Boltzmann probability distribution is used as a model to identify network parameters by optimizing the likelihood of predicting an output given hidden states trained on available data. Training such networks often requires sampling over a large probability space that must be approximated during gradient based optimization. Quantum annealing has been proposed as a means to search this space more efficiently which has been experimentally investigated on D-Wave hardware. D-Wave implementation requires selection of an effective inverse temperature or hyperparameter (β) within the Boltzmann distribution which can strongly influence optimization. Here, we show how this parameter can be estimated as a hyperparameter applied to D-Wave hardware during neural network training by maximizing the likelihood or minimizing the Shannon entropy. We find both methods improve training RBMs based upon D-Wave hardware experimental validation on an image recognition problem. Neural network image reconstruction errors are evaluated using Bayesian uncertainty analysis which illustrate more than an order magnitude lower image reconstruction error using the maximum likelihood over manually optimizing the hyperparameter. The maximum likelihood method is also shown to out-perform minimizing the Shannon entropy for image reconstruction.
97 MATHEMATICS AND COMPUTING↗
Leveraging In-Network Computing and Programmable Switches for Streaming Analysis of Scientific Data
With the emergence of programmable network devices that match the performance of fixed function devices, several recent projects have explored in-network computing, where the processing that is traditionally done outside the network is offloaded to the network devices. In-network computing has typically been applied to network functions (e.g., load balancing, NAT, and DNS), caching, data reduction/aggregation, and coordination/consensus functions. In some cases it has been used to accelerate stream-processing tasks that involve small payloads and simple operations. In this work we focus on leveraging in-network computing for stream processing of scientific datasets with large payloads that require complex operations such as floating-point computations and logarithmic functions. We demonstrate in-network computing for a real-world scientific application performing streaming normalization of a 2-D image from a light source experiment. We discuss the challenges we encountered and potential approaches to address them.
Sankaran, Ganesh↗
Low Power, Radiation Resilient Synchronous Edge Processing for Remote Monitoring
Next-generation space remote sensing systems may be equipped with imaging arrays that sense data at a rate that outstrips the processing capability of any computing hardware that can operate within a satellite’s power budget. This project developed novel convolutional and recurrent neural networks to detect and estimate point-like events amid clutter, and investigated their efficient and accurate implementation on analog in-memory computing systems that are 10-1000× more energy-efficient than digital processors. This project leveraged two memory devices at different levels of technological maturity: a large-scale analog computing prototype using commercial SONOS charge-trap memory, and electrochemical memory (ECRAM) with intrinsic radiation hardness. We experimentally demonstrated end-to-end analog processing of our neural networks on SONOS and characterized the radiation response of both SONOS and ECRAM. We advanced the state-of-the-art in ECRAM precision and reliability, and developed co-design methods to enable accurate long-term operation of SONOS analog accelerators in space radiation environments.
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗