Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Feature Extraction”

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 703 records · Page 39
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Distributed non-negative matrix factorization with determination of the number of latent features

The holistic analysis and understanding of the latent (that is, not directly observable) variables and patterns buried in large datasets is crucial for data-driven science, decision making and emergency response. Such exploratory analyses require devising unsupervised learning methods for data mining and extraction of the latent features, and non-negative matrix factorization (NMF) is one of the prominent such methods. NMF is based on compute-intense non-convex constrained minimization, which, for large datasets requires fast and distributed algorithms. However, current parallel implementations of NMF fail to estimate the number of latent features. In practice, identifying these features is both difficult and significant for pattern recognition and latent feature analysis, especially for large dense matrices. Here, we introduce a distributed NMF algorithm coupled with distributed custom clustering followed by a stability analysis on dense data, which we call DnMFk, to determine the number of latent variables. The results on synthetic data and the classical Swimmer data set demonstrate the accuracy of model determination while scaling nearly linearly across multiple processors for large data. Further, we employ DnMFk to determine the number of hidden features from a terabyte matrix.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Learning View-Invariant Features for Person Identification in Temporally Synchronized Videos Taken by Wearable Cameras

In this paper, we study the problem of Cross-View Person Identification (CVPI), which aims at identifying the same person from temporally synchronized videos taken by different wearable cameras. Our basic idea is to utilize the human motion consistency for CVPI, where human motion can be computed by optical flow. However, optical flow is view-variant - the same person's optical flow in different videos can be very different due to view angle change. In this paper, we attempt to utilize 3D human-skeleton sequences to learn a model that can extract view-invariant motion features from optical flows in different views. For this purpose, we use 3D Mocap database to build a synthetic optical flow dataset and train a Triplet Network (TN) consisting of three sub-networks: two for optical flow sequences from different views and one for the underlying 3D Mocap skeleton sequence. Finally, sub-networks for optical flows are used to extract view-invariant features for CVPI. Experimental results show that, using only the motion information, the proposed method can achieve comparable performance with the state-of-the-art methods. Further combination of the proposed method with an appearance-based method achieves new state-of-the-art performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Applications of Machine Learning to Predicting Core-collapse Supernova Explosion Outcomes

Most existing criteria derived from progenitor properties of core-collapse supernovae are not very accurate in predicting explosion outcomes. We present a novel look at identifying the explosion outcome of core-collapse supernovae using a machine-learning approach. Informed by a sample of 100 2D axisymmetric supernova simulations evolved with F ornax , we train and evaluate a random forest classifier as an explosion predictor. Furthermore, we examine physics-based feature sets including the compactness parameter, the Ertl condition, and a newly developed set that characterizes the silicon/oxygen interface. With over 1500 supernovae progenitors from 9-27 M ⊙ , we additionally train an autoencoder to extract physics-agnostic features directly from the progenitor density profiles. We find that the density profiles alone contain meaningful information regarding their explodability. Both the silicon/oxygen and autoencoder features predict the explosion outcome with ≈90% accuracy. In anticipation of much larger multidimensional simulation sets, we identify future directions in which machine-learning applications will be useful beyond the explosion outcome prediction.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Domain Adaptive Graph Neural Networks for Constraining Cosmological Parameters Across Multiple Data Sets

Deep learning models have been shown to outperform methods that rely on summary statistics, like the power spectrum, in extracting information from complex cosmological data sets. However, due to differences in the subgrid physics implementation and numerical approximations across different simulation suites, models trained on data from one cosmological simulation show a drop in performance when tested on another. Similarly, models trained on any of the simulations would also likely experience a drop in performance when applied to observational data. Training on data from two different suites of the CAMELS hydrodynamic cosmological simulations, we examine the generalization capabilities of Domain Adaptive Graph Neural Networks (DA-GNNs). By utilizing GNNs, we capitalize on their capacity to capture structured scale-free cosmological information from galaxy distributions. Moreover, by including unsupervised domain adaptation via Maximum Mean Discrepancy (MMD), we enable our models to extract domain-invariant features. We demonstrate that DA-GNN achieves higher accuracy and robustness on cross-dataset tasks. Using data visualizations, we show the effects of domain adaptation on proper latent space data alignment. This shows that DA-GNNs are a promising method for extracting domain-independent cosmological information, a vital step toward robust deep learning for real cosmic survey data.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

A New Workflow of X-ray CT Image Processing and Data Analysis of Structural Features in Rock Using Open-Source Software

X-ray computed tomography (CT) images of rock specimens often contain artifacts which must be corrected before scientific analyses are performed. Here, we present a new workflow of automated image processing to utilize poor-quality X-ray CT scan images. The workflow runs on the open-source image analysis software and efficiently separates desired features from low-contrast scanned images. The new workflow is a two-step technique using contrast enhancement and automated feature segmentation to generate noise-free binary images. The results of binary images using the proposed workflow and using a conventional thresholding technique are analyzed to show the quality of the proposed method. The paper also presents a workflow of estimating the structural geometries of features in two and three dimensions. The results of the structural feature analyses and computational time were compared between the open-source (ImageJ) and commercial image analysis software (Bruker Computed Tomography Analyzer). The commercial software was more computationally efficient, but the task-specific macros in open-source software enabled the user-desired automation in image processing and data extraction of desired structural features of comparable quality.

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

The Near-infrared Spectrum of the Nuclear Star Cluster: Looking below the Tip of the Iceberg, and Comparisons with Extragalactic Nuclei

Long-slit near-infrared (NIR) spectra of the Galactic nuclear star cluster (NSC) are discussed. The spectra sample the major axis of the NSC out to its half-light radius. The absorption spectrum of the central regions of the NSC is averaged over angular scales of tens of arc seconds in order to sample globular cluster-like total luminosities, and the results are compared with model spectra. The equivalent widths of Na i2.21 μm and Ca 2.26 μm outside of the center of the NSC, where light from nuclear bulge stars contributes a large fraction to the total flux, are consistent with solar chemical mixtures. In contrast, the equivalent widths of Na i2.21 μm and Ca i2.26 μm near the center of the NSC are larger than expected from models with solar chemical mixtures, even after light from the brightest evolved stars is removed. The depths of spectroscopic features changing along the major axis of the NSC are consistent with imaging studies that have found evidence of population gradients in the NSC. That Na i2.21 μm and Ca i2.26 μm are deeper than predicted for solar chemical mixtures over a range of evolutionary states is consistent with previous studies that find that the majority of stars near the center of the NSC formed from material that had nonsolar chemical mixtures. The depths of the Na i2.21 μm and Ca i2.26 μm features in the central regions of the NSC are comparable to those in the nuclear spectrum of the early-type Virgo disk galaxy NGC 4491, and are deeper than in the central spectra of NGC 253 and 7793. A spectrum of nebular emission and the youngest stars near the galactic center is also extracted. The equivalent widths of emission features in the extracted NIR spectrum are similar to those in the nuclear spectrum of NGC 253, and it is argued that this agreement is best achieved if the current episode of star formation near the center of the NSC has been in progress for at least a few megayears.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

Lensing Constraints on the Mass Profile Shape and the Splashback Radius of Galaxy Clusters

The lensing signal around galaxy clusters can, in principle, be used to test detailed predictions for their average mass profile from numerical simulations. However, the intrinsic shape of the profiles can be smeared out when a sample that spans a wide range of cluster masses is averaged in physical length units. This effect especially conceals rapid changes in gradient such as the steep drop associated with the splashback radius, a sharp edge corresponding to the outermost caustic in accreting halos. We optimize the extraction of such local features by scaling individual halo profiles to a number of spherical overdensity radii, and apply this method to 16 X-ray-selected, high-mass clusters targeted in the Cluster Lensing And Supernova survey with Hubble. By forward-modeling the weak- and strong-lensing data presented by Umetsu et al., we show that, regardless of the scaling overdensity, the projected ensemble density profile is remarkably well described by a Navarro–Frenk–White (NFW) or Einasto profile out to R∼2.5 h{sup −1} Mpc, beyond which the profiles flatten. We constrain the NFW concentration to c{sub 200c}=3.66±0.11 at M{sub 200c}≃1.0×10{sup 15} h{sup −1} M{sub ⊙}, consistent with and improved from previous work that used conventionally stacked lensing profiles, and in excellent agreement with theoretical expectations. Assuming the profile form of Diemer and Kravtsov and generic priors calibrated from numerical simulations, we place a lower limit on the splashback radius of the cluster halos, if it exists, of R{sub sp}{sup 3D}/r{sub 200m}>0.89 (R{sub sp}{sup 3D}>1.83 h{sup −1} Mpc) at 68% confidence. The corresponding density feature is most pronounced when the cluster profiles are scaled by r{sub 200m}, and smeared out when scaled to higher overdensities.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Unsupervised Power System Event Detection and Classification Using Unlabeled PMU Data

This paper proposes a novel data-driven power system event detection and classification method based on 5TB of actual PMU measurements collected from the US western interconnect. Firstly, a set of comprehensive power quality rules are proposed to pre-filter the raw data and extract the regions of interest (ROI). Six distinct event categories are defined and corresponding patterns are chosen as references. Meanwhile, detailed characteristics of patterns are summarized to enhance our understanding of the actual events. Then, the time-independent feature vectors are generated by extracting the statistical, temporal, and spectral features from the raw time-series data. Furthermore, an ensemble model is proposed to cluster the events by combining multiple K-means clustering models using a voting strategy. Besides, both system-level and PMU-level clustering models are developed. The accuracy and robustness of the event detection method are further improved through interactive evaluation of the two-level clustering results. This paper summarizes the actual characteristics of each event category and provides a reliable basis for accurate label generation. The experiments demonstrate the effectiveness of the proposed event detection and classification method.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2021

Unsupervised Power System Event Detection and Classification Using Unlabeled PMU Data

This paper proposes a novel data-driven power system event detection and classification method based on 5TB of actual PMU measurements collected from the US western interconnect. Firstly, a set of comprehensive power quality rules are proposed to pre-filter the raw data and extract the regions of interest (ROI). Six distinct event categories are defined and corresponding patterns are chosen as references. Meanwhile, detailed characteristics of patterns are summarized to enhance our understanding of the actual events. Then, the time-independent feature vectors are generated by extracting the statistical, temporal, and spectral features from the raw time-series data. Furthermore, an ensemble model is proposed to cluster the events by combining multiple K-means clustering models using a voting strategy. Besides, both system-level and PMU-level clustering models are developed. The accuracy and robustness of the event detection method are further improved through interactive evaluation of the two-level clustering results. This paper summarizes the actual characteristics of each event category and provides a reliable basis for accurate label generation. The experiments demonstrate the effectiveness of the proposed event detection and classification method.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

Glass Refraction Distortion Object Detection via Abstract Features

Glass reflection and refraction lead to missing and distorted object feature data, affecting the accuracy of object detection. In order to solve the above problems, this paper proposed a glass refraction distortion object detection via abstract features. The number of parameters of the algorithm is reduced by introducing skip connections and expansion modules with different expansion rates. The abstract feature information of the object is extracted by binary cross-entropy loss. Meanwhile, the abstract feature distance between the object domain and source domain is reduced by a loss function, which improves the accuracy of object detection under glass interference. To verify the effectiveness of the algorithm in this paper, the GRI dataset is produced and made public on GitHub. The algorithm of this paper is compared with the current state-of-the-art Deep Face, VGG Face, TBE-CNN, DA-GAN, PEN-3D, LMZMPM, and the average detection accuracy of our algorithm is 92.57% at the highest, and the number of parameters is only 5.13 M.

Cai, Lei↗
DOE OSTI2023

Bridging the Gap between Cosmological Simulations with Graph Neural Networks and Domain Adaptation

Deep learning models have been shown to outperform methods that rely on summary statistics, like the power spectrum, in extracting information from complex cosmological data sets. However, due to differences in the subgrid physics implementation and numerical approximations across different simulation suites, models trained on data from one cosmological simulation show a drop in performance when tested on another. Similarly, models trained on any of the simulations would also likely experience a drop in performance when applied to observational data. Training on data from two different suites of the CAMELS hydrodynamic cosmological simulations, we examine the generalization capabilities of Domain Adaptive Graph Neural Networks (DA-GNNs). By utilizing GNNs, we capitalize on their capacity to capture structured scale-free cosmological information from galaxy distributions. Moreover, by including unsupervised domain adaptation via Maximum Mean Discrepancy (MMD), we enable our models to extract domain-invariant features. We demonstrate that DA-GNN achieves higher accuracy and robustness on cross dataset tasks (up to 28% better relative error and up to almost an order of magnitude better χ 2 ). Using data visualizations, we show the effects of domain adaptation on proper latent space data alignment. This shows that DA-GNNs are a promising method for extracting domain-independent cosmological information, a vital step toward robust deep learning for real cosmic survey data.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Improving deep learning performance for predicting large-scale geological ${{CO}_{2}}$ sequestration modeling through feature coarsening

Physics-based reservoir simulation for fluid flow in porous media is a numerical simulation method to predict the temporal-spatial patterns of state variables (e.g. pressure p) in porous media, and usually requires prohibitively high computational expense due to its non-linearity and the large number of degrees of freedom (DoF). This work describes a deep learning (DL) workflow to predict the pressure evolution as fluid flows in large-scale 3-dimensional(3D) heterogeneous porous media. In particular, we develop an efficient feature coarsening technique to extract the most representative information and perform the training and prediction of DL at the coarse scale, and further recover the resolution at the fine scale by spatial interpolation. We validate the DL approach to predict pressure field against physics-based simulation data for a field-scale 3D geologic CO 2 sequestration reservoir model. We evaluate the impact of feature coarsening on DL performance, and observe that the feature coarsening not only decreases the training time by >74% and reduces the memory consumption by >75%, but also maintains temporal error 0.63% on average. Besides, the DL workflow provides predictive efficiency with 1406 times speedup compared to physics-based numerical simulation. The key findings from this research significantly improve the training and prediction efficiency of deep learning model to deal with large-scale heterogeneous reservoir models, and thus it can also be further applied to accelerate workflows of history matching and reservoir optimization for close-loop reservoir management.

58 GEOSCIENCES↗
DOE OSTI2023

Domain Adaptive Graph Neural Networks for Constraining Cosmological Parameters Across Multiple Data Sets

Deep learning models have been shown to outperform methods that rely on summary statistics, like the power spectrum, in extracting information from complex cosmological data sets. However, due to differences in the subgrid physics implementation and numerical approximations across different simulation suites, models trained on data from one cosmological simulation show a drop in performance when tested on another. Similarly, models trained on any of the simulations would also likely experience a drop in performance when applied to observational data. Training on data from two different suites of the CAMELS hydrodynamic cosmological simulations, we examine the generalization capabilities of Domain Adaptive Graph Neural Networks (DA-GNNs). By utilizing GNNs, we capitalize on their capacity to capture structured scale-free cosmological information from galaxy distributions. Moreover, by including unsupervised domain adaptation via Maximum Mean Discrepancy (MMD), we enable our models to extract domain-invariant features. We demonstrate that DA-GNN achieves higher accuracy and robustness on cross-dataset tasks (up to $28\%$ better relative error and up to almost an order of magnitude better $\chi^2$). Using data visualizations, we show the effects of domain adaptation on proper latent space data alignment. This shows that DA-GNNs are a promising method for extracting domain-independent cosmological information, a vital step toward robust deep learning for real cosmic survey data.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

SU-D-BRA-04: Fractal Dimension Analysis of Edge-Detected Rectal Cancer CTs for Outcome Prediction

Purpose: To extract the fractal dimension features from edge-detected rectal cancer CTs, and to examine the predictability of fractal dimensions to outcomes of primary rectal cancer patients. Methods: Ninety-seven rectal cancer patients treated with neo-adjuvant chemoradiation were enrolled in this study. CT images were obtained before chemoradiotherapy. The primary lesions of the rectal cancer were delineated by experienced radiation oncologists. These images were extracted and filtered by six different Laplacian of Gaussian (LoG) filters with different filter values (0.5–3.0: from fine to coarse) to achieve primary lesions in different anatomical scales. Edges of the original images were found at zero-crossings of the filtered images. Three different fractal dimensions (box-counting dimension, Minkowski dimension, mass dimension) were calculated upon the image slice with the largest cross-section of the primary lesion. The significance of these fractal dimensions in survival, recurrence and metastasis were examined by Student’s t-test. Results: For a follow-up time of two years, 18 of 97 patients had experienced recurrence, 24 had metastasis, and 18 were dead. Minkowski dimensions under large filter values (2.0, 2.5, 3.0) were significantly larger (p=0.014, 0.006, 0.015) in patients with recurrence than those without. For metastasis, only box-counting dimensions under a single filter value (2.5) showed differences (p=0.016) between patients with and without. For overall survival, box-counting dimensions (filter values = 0.5, 1.0, 1.5), Minkowski dimensions (filter values = 0.5, 1.5, 2.0, 2,5) and mass dimensions (filter values = 1.5, 2.0) were all significant (p<0.05). Conclusion: It is feasible to extract shape information by edge detection and fractal dimensions analysis in neo-adjuvant rectal cancer patients. This information can be used to prognosis prediction.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

Characterizing the DNA damage response by cell tracking algorithms and cell features classification using high-content time-lapse analysis

Traditionally, the kinetics of DNA repair have been estimated using immunocytochemistry by labeling proteins involved in the DNA damage response (DDR) with fluorescent markers in a fixed cell assay. However, detailed knowledge of DDR dynamics across multiple cell generations cannot be obtained using a limited number of fixed cell time-points. Here we report on the dynamics of 53BP1 radiation induced foci (RIF) across multiple cell generations using live cell imaging of non-malignant human mammary epithelial cells (MCF10A) expressing histone H2B-GFP and the DNA repair protein 53BP1-mCherry. Using automatic extraction of RIF imaging features and linear programming techniques, we were able to characterize detailed RIF kinetics for 24 hours before and 24 hours after exposure to low and high doses of ionizing radiation. High-content-analysis at the single cell level over hundreds of cells allows us to quantify precisely the dose dependence of 53BP1 protein production, RIF nuclear localization and RIF movement after exposure to X-ray. Using elastic registration techniques based on the nuclear pattern of individual cells, we could describe the motion of individual RIF precisely within the nucleus. We show that DNA repair occurs in a limited number of large domains, within which multiple small RIFs form, merge and/or resolve with random motion following normal diffusion law. Large foci formation is shown to be mainly happening through the merging of smaller RIF rather than through growth of an individual focus. We estimate repair domain sizes of 7.5 to 11 µm 2 with a maximum number of ~15 domains per MCF10A cell. This work also highlights DDR which are specific to doses larger than 1 Gy such as rapid 53BP1 protein increase in the nucleus and foci diffusion rates that are significantly faster than for spontaneous foci movement. We hypothesize that RIF merging reflects a "stressed" DNA repair process that has been taken outside physiological conditions when too many DSB occur at once. High doses of ionizing radiation lead to RIF merging into repair domains which in turn increases DSB proximity and misrepair. Furthermore, such finding may therefore be critical to explain the supralinear dose dependence for chromosomal rearrangement and cell death measured after exposure to ionizing radiation.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

An Explainable Classification Framework for Determining and Understanding the Suitability of Solvent Extraction for Bioproduct Recovery

Lignocellulosic biomass is an abundant feedstock for producing sustainable fuels and chemicals. However, a key challenge in most biomass utilization strategies is the recovery of products from a dilute, typically aqueous, phase. In this respect, liquid–liquid extraction, which relies on a solvent to transfer a product of interest from one liquid phase to another (solvent-rich) phase, is a technology that can reduce the energy requirements for product recovery. To reduce solvent consumption, liquid–liquid extraction needs to be combined with another separation method (e.g., distillation) to recycle the solvent. Despite the research on solvent extraction, there are limited system-wide methods that allow us to determine when extraction is well suited to carry out a specific separation. Accordingly, we present a classification framework to predict whether extraction, coupled with distillation, is feasible and more economical than distillation. Our framework is based on features such as feed composition, liquid–liquid equilibrium constants, relative volatilities, and solvent price, and leads to trained classifiers that show good prediction accuracy. We further study how specific features influence the suitability of extraction. Furthermore, to showcase the applicability of the framework, we use it to analyze the separation of acetic acid from water.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2018

Repeatability and Reproducibility of Radiomic Features: A Systematic Review

An ever-growing number of predictive models used to inform clinical decision making have included quantitative, computer-extracted imaging biomarkers, or “radiomic features.” Broadly generalizable validity of radiomics-assisted models may be impeded by concerns about reproducibility. We offer a qualitative synthesis of 41 studies that specifically investigated the repeatability and reproducibility of radiomic features, derived from a systematic review of published peer-reviewed literature.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2022

ThunderSecure: deploying real-time intrusion detection for 100G research networks by leveraging stream-based features and one-class classification network

Nowadays, data generated by large-scale scientific experiments are on the scale of petabytes per month. These data are transferred through dedicated high-bandwidth networks (40/100G) across distributed sites for processing, storage, and analysis. Like general purpose networks, research networks experience intrusions. However, monitoring anomalies in such high-speed network traffics is challenging given current cyber-infrastructure. Moreover, traditional network intrusion detection systems (NIDS) are signature based. However, anomaly patterns are difficult to define and that rulesets are often not updated frequently enough to reflect the changes of attack behaviors. We present ThunderSecure, a high-throughput, unsupervised learning-based intrusions detection system for 100G research networks. ThunderSecure implements an efficient packet processing and detection pipeline using multi-cores and GPUs. It extracts statistical and temporal features from real-time network data streams and feeds them to a one-class anomaly detection network. A baseline of normal distribution will be created based on the training observation. Testing traffic deviated from the learned profile will be marked as anomalies. We trained ThunderSecure on hundreds of billions of science data packets mirrored from two 100G network connections at Fermi National Accelerator Laboratory. The detection performance was evaluated on traffic captured from the same research network days and weeks after the training with different types of attack flows injected. Results show that ThunderSecure can recognize science data traffic captured long after the training and made nearly certain detection on the segment of the streams where anomalous flows were injected.

100G research network↗