Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep data”

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 1,135 records · Page 63
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Detecting Low Surface Brightness Galaxies with Mask R-CNN

Low surface brightness galaxies (LSBGs), galaxies that are fainter than the dark night sky, are famously difficult to detect. However, studies of these galaxies are essential to improve our understanding of the formation and evolution of low-mass galaxies. In this work, we train a deep learning model using the Mask R-CNN framework on a set of simulated LSBGs inserted into images from the Dark Energy Survey (DES) Data Release 2 (DR2). This deep learning model is combined with several conventional image pre-processing steps to develop a pipeline for the detection of LSBGs. We apply this pipeline to the full DES DR2 coadd image dataset, and preliminary results show the detection of 22 large, high-quality LSBG candidates that went undetected by conventional algorithms. Furthermore, we find that Galactic cirrus represents the largest contaminant in our resulting candidate list.

Levy, Caleb↗
DOE OSTI2020

Chapter 14: Machine Learning of Combustion LES Models from Reacting Direct Numerical Simulation

In this chapter we demonstrate how supervised deep learning techniques can be used to construct models for the filtered progress variable source term necessary for large eddy simulation (LES). The source data for the model is a direct numerical simulation (DNS) of a reacting flow in a low swirl burner configuration. Filtered quantities taken from the DNS data are used to train a deep neural network (DNN)-based model. An efficient data sampling strategy was devised to ensure that a uniform representation of all the states observed in the filtered DNS data are equally present in the training dataset. A-priori testing of the DNN-based model highlights the representative power of DNN to accurately reproduce the filtered reaction progress variable source term over a range of scales and various flame regimes as seen in an industrial burner.

combustion LES models↗
DOE OSTI2022

Scalable algorithms for physics-informed neural and graph networks

Physics-informed machine learning (PIML) has emerged as a promising new approach for simulating complex physical and biological systems that are governed by complex multiscale processes for which some data are also available. In some instances, the objective is to discover part of the hidden physics from the available data, and PIML has been shown to be particularly effective for such problems for which conventional methods may fail. Unlike commercial machine learning where training of deep neural networks requires big data, in PIML big data are not available. Instead, we can train such networks from additional information obtained by employing the physical laws and evaluating them at random points in the space–time domain. Such PIML integrates multimodality and multifidelity data with mathematical models, and implements them using neural networks or graph networks. Here, we review some of the prevailing trends in embedding physics into machine learning, using physics-informed neural networks (PINNs) based primarily on feed-forward neural networks and automatic differentiation. For more complex systems or systems of systems and unstructured data, graph neural networks (GNNs) present some distinct advantages, and here we review how physics-informed learning can be accomplished with GNNs based on graph exterior calculus to construct differential operators; we refer to these architectures as physics-informed graph networks (PIGNs). We present representative examples for both forward and inverse problems and discuss what advances are needed to scale up PINNs, PIGNs and more broadly GNNs for large-scale engineering problems.

42 ENGINEERING↗
NASA NTRS2016

Satellite-Based Spatiotemporal Trends in PM2.5 Concentrations: China 2004-2013

Three decades of rapid economic development is causing severe and widespread PM2.5(particulate matter (is) less than 2.5 ) pollution in China. However, research on the health impacts of PM2.5 exposure has been hindered by limited historical PM2.5 concentration data. We estimated ambient PM2.5 concentrations from 2004 to 2013 in China at 0.1 deg resolution using the most recent satellite data and evaluated model performance with available ground observations. We developed a two-stage spatial statistical model using the Moderate Resolution Imaging Spectroradiometer (MODIS) Collection 6 aerosol optical depth (AOD) and assimilated meteorology, land use data, and PM2.5 concentrations from China's recently established ground monitoring network. An inverse variance weighting (IVW) approach was developed to combine MODIS Dark Target and Deep Blue AOD to optimize data coverage. We evaluated model predicted PM2.5 concentrations from 2004 to early 2014 using ground observations. The overall model cross-validation R(sup 2) and relative prediction error were 0.79 and 35.6%, respectively. Validation beyond the model year (2013) indicated that it accurately predicted PM(sub 2.5) concentrations with little bias at the monthly (R(sup 2) = 0.73), regression slope = 0.91) and seasonal (R(sup 2) = 0.79), regression slope = 0.92) levels. Seasonal variations revealed that winter was the most polluted season and that summer was the cleanest season. Analysis of predicted PM2.5 levels showed a mean annual increase of 1.97 micro-g/cu cm between 2004 and 2007 and a decrease of 0.46 micro-g/cu cm between 2008 and 2013. Our satellite-driven model can provide reliable historical PM2.5 estimates in China at a resolution comparable to those used in epidemiologic studies on the health effects of long-term PM2.5 exposure in North America. This data source can potentially advance research on PM2.5 health effects in China.

satellite remote sensing↗
DOE OSTI2018

End-to-End Event Classification of High-Energy Physics Data

Feature extraction algorithms, such as convolutional neural networks, have introduced the possibility of using deep learning to train directly on raw data without the need for rule-based feature engineering. In the context of particle physics, such end-to-end approaches can be used for event classification to learn directly from detector-level data in a way that is completely independent of the high-level physics reconstruction. We demonstrate a technique for building such end-to-end event classifiers to distinguish simulated electromagnetic decays in a high-fidelity model of the CMS Electromagnetic Calorimeter.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

Intelligent Monitoring System for Real-time Geologic Storage, Optimization, and Reservoir Management

The objective of the subtask was to develop a near-real-time monitoring system for seismic data at the Decatur, IL, geologic carbon sequestration (GCS) site and specifically include fiber-optic cable derived distributed acoustic signal (DAS) data in the process. Owing to the large volumes of data, we opted to utilize existing deep borehole conventional seismic sensors for detection and pull DAS and shallow borehole seismic data once a detection has been made. Unfortunately, the horizontal fiber-optic cables did not yield microseismic signals for use in locating events near the GCS site. Various stacking and filtering approaches were tested without any coherent detection becoming apparent. We attribute the insensitivity to local microseismic events to a lack of coupling in downhole, vertical cable and the non-ideal alignment of the horizontal fiber-optic cable to the vertically polarized seismic energy. Despite the inability to detect and utilize the DAS data from the fiber-optic cables, we developed a general processing framework that enables easy adaptation for future deployment of fiber-optic cables with better suited alignment at Decatur or elsewhere.

Kaven, J. Ole↗
NASA NTRS1984

Evidence for slow-mode shocks in the deep geomagnetic tail

Plasma and field data from the ISEE-3 space probe provide evidence that the lobe-plasma sheet boundary is the central part of the distant geomagnetic tail is often a slow-mode shock. Such shocks are predicted by Petchek's (1964) model of the reconnection. If this model applies, then the ISEE-3 observations place the general location of the reconnection diffusion region closer to the earth than x of about -100 earth radii.

Feldman, W. C.↗
DOE OSTI2022

Deep-learning-guided high-resolution subsurface reflectivity imaging with application to ground-penetrating radar data

Subsurface reflectivity imaging is one of the most important geophysical characterization methods for revealing subsurface structures. In many cases, accurate subsurface reflectivity imaging is challenging because of, for example, random or coherent noise in the data and sparse source-receiver observation geometry. Here, we develop a deep-learning-guided iterative imaging method to improve subsurface structure imaging. Specifically, we train a supervised neural network to infer a noise-free, high-resolution image from a noisy, low-resolution image and use this estimated image as guidance to regularize least-squares imaging. We develop a systematic method to generate high-quality synthetic training data (data-label pairs) to train the guidance neural network. The trained neural network can provide high-fidelity predictions even for field-data images that are not in the training data. We validate our new imaging method using one synthetic and two field ground-penetrating radar data examples, and find that our method can produce clean, high-resolution subsurface reflectivity images where existing single-pass and least-squares imaging methods fail due to noise and insufficient data coverage.

58 GEOSCIENCES↗
NASA NTRS1986

Energetic charged particles in the Uranian magnetosphere

The cosmic ray instrument on Voyager 2 obtained detailed charged particle data on the path the spacecraft followed through the newly-discovered Uranian magnetosphere. Notably, the electron counts did not rise above background until the spacecraft was deep in the magnetosphere. The data further indicated that the satellites Miranda, Ubriel and Ariel impose a sweep effect on MeV electrons, which move radially inward from some source in the outer magnetosphere or the magnetotail. The energetic particle data also supported a firm pitch angle dependence for proton energy spectra in the 1-8 MeV range. Finally, the particle density data were commensurate with a dipole magnetic field tilted 60.1 deg in relation to the planet and rotating every 17.4 hr, close to estimates from other instruments.

Stone, E. C.↗
NASA NTRS2005

Response to MRO's end-to-end data accountability challenges

(MRO) on August 12, 2005. It carries six science instruments and three engineering payloads. Because MRO will produce an unprecedented number of science products, it will transmit a much higher data volume via high data rate than any other deep space mission to date. Keeping track of MRO products as well as relay products would be a daunting, expensive task without a well-planned data-product tracking strategy. To respond to this challenge, the MRO project developed the End-to- End Data Accountability System by utilizing existing information available from both ground and flight elements. Therefore, a capability to perform first-order problem diagnosis is essential in order for MRO to answer the questions, where is my data? and when will my data be available? This paper details the approaches taken, design and implementation of the tools, procedures and teams that track data products from the time they are predicted until they arrive in the hands of the end users.

data delivery status tracking↗
NASA NTRS2008

On Certain New Methodology for Reducing Sensor and Readout Electronics Circuitry Noise in Digital Domain

NASA Hubble Space Telescope (HST) and upcoming cosmology science missions carry instruments with multiple focal planes populated with many large sensor detector arrays. These sensors are passively cooled to low temperatures for low-level light (L3) and near-infrared (NIR) signal detection, and the sensor readout electronics circuitry must perform at extremely low noise levels to enable new required science measurements. Because we are at the technological edge of enhanced performance for sensors and readout electronics circuitry, as determined by thermal noise level at given temperature in analog domain, we must find new ways of further compensating for the noise in the signal digital domain. To facilitate this new approach, state-of-the-art sensors are augmented at their array hardware boundaries by non-illuminated reference pixels, which can be used to reduce noise attributed to sensors. There are a few proposed methodologies of processing in the digital domain the information carried by reference pixels, as employed by the Hubble Space Telescope and the James Webb Space Telescope Projects. These methods involve using spatial and temporal statistical parameters derived from boundary reference pixel information to enhance the active (non-reference) pixel signals. To make a step beyond this heritage methodology, we apply the NASA-developed technology known as the Hilbert- Huang Transform Data Processing System (HHT-DPS) for reference pixel information processing and its utilization in reconfigurable hardware on-board a spaceflight instrument or post-processing on the ground. The methodology examines signal processing for a 2-D domain, in which high-variance components of the thermal noise are carried by both active and reference pixels, similar to that in processing of low-voltage differential signals and subtraction of a single analog reference pixel from all active pixels on the sensor. Heritage methods using the aforementioned statistical parameters in the digital domain (such as statistical averaging of the reference pixels themselves) zeroes out the high-variance components, and the counterpart components in the active pixels remain uncorrected. This paper describes how the new methodology was demonstrated through analysis of fast-varying noise components using the Hilbert-Huang Transform Data Processing System tool (HHT-DPS) developed at NASA and the high-level programming language MATLAB (Trademark of MathWorks Inc.), as well as alternative methods for correcting for the high-variance noise component, using an HgCdTe sensor data. The NASA Hubble Space Telescope data post-processing, as well as future deep-space cosmology projects on-board instrument data processing from all the sensor channels, would benefit from this effort.

Kizhner, Semion↗
DOE OSTI2024

Missing Wedge Completion via Unsupervised Learning with Coordinate Networks

Cryogenic electron tomography (cryoET) is a powerful tool in structural biology, enabling detailed 3D imaging of biological specimens at a resolution of nanometers. Despite its potential, cryoET faces challenges such as the missing wedge problem, which limits reconstruction quality due to incomplete data collection angles. Recently, supervised deep learning methods leveraging convolutional neural networks (CNNs) have considerably addressed this issue; however, their pretraining requirements render them susceptible to inaccuracies and artifacts, particularly when representative training data is scarce. To overcome these limitations, we introduce a proof-of-concept unsupervised learning approach using coordinate networks (CNs) that optimizes network weights directly against input projections. This eliminates the need for pretraining, reducing reconstruction runtime by 3–20× compared to supervised methods. Our in silico results show improved shape completion and reduction of missing wedge artifacts, assessed through several voxel-based image quality metrics in real space and a novel directional Fourier Shell Correlation (FSC) metric. Our study illuminates benefits and considerations of both supervised and unsupervised approaches, guiding the development of improved reconstruction strategies.

42 ENGINEERING↗
DOE OSTI2018

Automated Labeling of Electron Microscopy Images Using Deep Learning

© 2018 IEEE. Searching for scientific data requires metadata providing a relevant context. Today, generating metadata is a time and labor intensive manual process that is often neglected, and important datasets are not accessible through search. We investigate the use of machine learning to generalize metadata from a subset of labeled data, thus increasing the availability of meaningful metadata for search. Specifically, we consider electron microscopy images collected at the National Center for Electron Microscopy (NCEM) at the Lawrence Berkeley National Laboratory (LBNL) and use deep learning to discern characteristics from a small subset of labeled images and transfer labels to the entire image corpus. Relatively small training set sizes and a minimum resolution of 512\times 512 pixels required by the application domain pose unique challenges. We overcome these challenges by using a simple yet powerful convolutional network architecture that limits the number of free parameters to lower the required amount of computational power and reduce the risk of overfitting. We achieve a classification accuracy of approximately 80% in discerning between images recorded in two operating modes of the electron microscope-transmission electron microscopy (TEM) and scanning transmission electron microscopy (STEM). We use transfer learning-i.e., re-using the pre-trained convolution layers from the TEM vs. STEM classification problem-to generalize labels and achieve an accuracy of approximately 70 % despite current experiments being limited to small training data sets. We present these predictions as suggestions to domain scientists through a web-based tool to accelerate the labeling process with the goal of further validating our approach and improving the accuracy of automatically created labels.

Weber, Gunther H↗
DOE OSTI2021

Development of a Deep Learning Model for Predicting the Drag Coefficients of Spherical and Non-Spherical Particles,

There is yet to be a well-established drag model for non-spherical particles required in a particle-laden flow that could cover a wide range of sphericities. This talk will explore the development of a general drag model for non-spherical particles by applying deep learning using available experimental data available in the literature. The integration of several raw experimental measurements from different sources and research directions allows the training of robust Artificial Intelligence and Machine Learning (ML) models. Neural networks are an ML approach inspired by the inner biological workings of the brain. This work aims to develop a Deep Neural Network (DNN) that predicts drag coefficient values with the ability to adapt appropriately to unseen data. Given the limited number of data points available and the variance found within the data collected from various sources, challenges may arise when looking to train the model. Our study tests and implements various model regularization techniques and assesses different loss and activation functions for the proposed DNN. The proposed model considers a broader range of features other than sphericity and Reynold number. These features include density ratio, solid volume fraction, lengthwise and crosswise sphericity, and more. Furthermore, we present the features that play a significant role in predicting different drag coefficients through feature importance. Within the investigated parameter ranges in this study, the following conclusions can be achieved and summarized below: • An improved drag coefficient model can be developed by considering more features such as, aspect ratio, lengthwise sphericity, crosswise sphericity, and density ratio. • DNN model can predict better results compared to traditional methods using MAE metric. • The proposed model addresses data challenges such as limited data and extreme data points through expanded feature-set and regularization. • Three major features that mostly affect the drag coefficient were identified from a feature importance analysis.

Presa-Reyes, Maria↗
DOE OSTI2025

Dataset, Code, and Models for Training Deep Learning Potentials for Low Temperature Plasma-Surface Interactions

This repository contains datasets, training scripts, and finished models, and test simulations used in the development of DeepREBO— a machine-learned interatomic potential trained to emulate the REBO2 empirical potential. The data was generated to study deep potential development for simulations of plasma-surface interactions. It uses an active learning framework, starting from a minimal dataset and iteratively expanding it. Included are those generated datasets, the trained models, and simulations used to evaluate the performance of the training process. This resource supports reproducibility and provides a reference framework for training deep potentials in plasma-surface interaction studies.

active learning↗
NASA NTRS2000

Participation In The AXAF Program As Interdisciplinary Scientist

We have supported the Advanced X-ray Astrophysics Facility (AXAF) project through specific studies and research in X-ray astronomy. Our research in X-ray astronomy has been concentrated in three projects: the ROSAT Deep Survey, the ROSAT All-sky Cluster Survey, and the study of moderate to high redshift clusters discovered serendipitously in long ROSAT pointings utilizing the wavelet transform. We have continued to develop the techniques and catalogues described in the publications listed below. The PI continued to participate in the SWG and attended its meetings except where conflicts existed. The Co-I supported the PI and attended all meetings in order to provide continuity. A post-doctoral fellow (Rosati) was hired to fully develop the techniques described above and make them applicable to the AXAF program. The results are described in the publications below. We continued to provide analysis of scientific and technical issues to the project scientist as appropriate. We made detailed simulations of the AXAF deep survey program and have made this available to the project and the ASC. The principal scientific results, discussed in great detail in the publications listed below, are as follows: (1) The wavelet transform is now a fine working tool for the analysis of faint X-ray sources especially clusters. (2) The analysis of the ROSAT DEEP CLUSTER SURVEY has given important constraints on cosmology particularly on the matter content of the universe and the amplitude of the fluctuations that created the galaxies and clusters. (3) The ROSAT DEEP SURVEY in the Lockman Hole has given an important'deep survey in the X-ray band that gives vital data on the distribution and evolution of X-ray sources. It is an important and complementary data base to compare with the AXAF deep fields. (4) The Northern ROSAT All Sky Cluster Survey now has a complete catalog and is again important and complementary to AXAF/ Chandra studies of clusters which are deeper but in a narrower area. These projects have had a significant impact on the whole field and have set up some excellent experiments to be undertaken with the AXAF/Chandra telescope such as the Chandra Deep Field South.

Giacconi, Riccardo↗
NASA NTRS2002

Nonaxisymmetric Variations Deep in the Convection Zone

Using a deep-focusing time-distance technique and the MDI medium-1 data, a preliminary study of nonaxisymmetric variability deep in the convection zone has been performed. The purpose of the present study is to see what signals might be present in raw travel times indicating variation. To this end, noise levels will be examined. Correlations with point separations in the range 40-50 deg. have been measured for the entire medium-1 dataset over a significant fraction of the solar disk. Both flows and mean-time variations have been examined. Separation of near-surface signals from deep signals will also be examined.

Duvall, Thomas↗