Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “FEATURE”

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 253 records · Page 14

Sparse Coding for N-Gram Feature Extraction and Training for File Fragment Classification

File fragment classification is an important step in the task of file carving in digital forensics. In file carving, files must be reconstructed based on their content as a result of their fragmented storage on disk or in memory. Existing methods for classification of file fragments typically use hand-engineered features such as byte histograms or entropy measures. In this paper, we propose an approach using sparse coding that enables automated feature extraction. Sparse coding, or sparse dictionary learning, is an unsupervised learning algorithm, and is capable of extracting features based simply on how well those features can be used to reconstruct the original data. With respect to file fragments, we learn sparse dictionaries for n-grams, continuous sequences of bytes, of different sizes. These dictionaries may then be used to estimate n-gram frequencies for a given file fragment, but for significantly larger n-gram sizes than are typically found in existing methods which suffer from combinatorial explosion. To demonstrate the capability of our sparse coding approach, we used the resulting features to train standard classifiers such as support vector machines (SVMs) over multiple file types. Experimentally, we achieved significantly better classification results with respect to existing methods, especially when the features were used in supplement to existing hand-engineered features.

97 MATHEMATICS AND COMPUTING↗

Basin-Scale Structural Features Database

The Basin-Scale Structural Features database provides spatial datasets of faults, fractures, folds, and earthquakes compiled from public, authoritative sources (e.g., U.S. Geological Survey and State Geological Surveys) and aggregated into derivative forms to support subsurface assessments. Recognizing that characterizing basin-scale structural features requires interpreting data that are often ambiguous or lack key information, the source data were evaluated using a knowledge-data framework and geospatial fuzzy logic method (Justman et al., 2020) to represent both measured (observed) and predicted (inferred or potential) structural features as derivative datasets. This workflow employs conceptual models for known structural features and predicted structural features, incorporating geospatial data to estimate potential, even with limited data. The aim is to aid and support an understanding of basin-scale features and identify potential gaps in data and knowledge. As of 4/30/2025, the database includes resources for nine sedimentary basins: Appalachian, Denver, U.S. Gulf Coast, Illinois, Michigan, Permian, Sacramento, San Joquin and Williston. The database is organized by basin and then data category: 1) Faults, fractures, folds, 2) Earthquakes, 3) Topographic, 4) Structural contours and isopachs, 5) Geophysical, and 6) Structural feature density assessment maps.

basin scale↗

Testing of Haar-Like Feature in Region of Interest Detection for Automated Target Recognition (ATR) System

The objectives of this project were to develop a ROI (Region of Interest) detector using Haar-like feature similar to the face detection in Intel's OpenCV library, implement it in Matlab code, and test the performance of the new ROI detector against the existing ROI detector that uses Optimal Trade-off Maximum Average Correlation Height filter (OTMACH). The ROI detector included 3 parts: 1, Automated Haar-like feature selection in finding a small set of the most relevant Haar-like features for detecting ROIs that contained a target. 2, Having the small set of Haar-like features from the last step, a neural network needed to be trained to recognize ROIs with targets by taking the Haar-like features as inputs. 3, using the trained neural network from the last step, a filtering method needed to be developed to process the neural network responses into a small set of regions of interests. This needed to be coded in Matlab. All the 3 parts needed to be coded in Matlab. The parameters in the detector needed to be trained by machine learning and tested with specific datasets. Since OpenCV library and Haar-like feature were not available in Matlab, the Haar-like feature calculation needed to be implemented in Matlab. The codes for Adaptive Boosting and max/min filters in Matlab could to be found from the Internet but needed to be integrated to serve the purpose of this project. The performance of the new detector was tested by comparing the accuracy and the speed of the new detector against the existing OTMACH detector. The speed was referred as the average speed to find the regions of interests in an image. The accuracy was measured by the number of false positives (false alarms) at the same detection rate between the two detectors.

neural network↗

Unsupervised Feature Learning for High-Resolution Satellite Image Classification

The rich data provided by high-resolution satellite imagery allow us to directly model geospatial neighborhoods by understanding their spatial and structural patterns. In this paper we explore an unsupervised feature learning approach to model geospatial neighborhoods for classification purposes. While pixel and object based classification approaches are widely used for satellite image analysis, often these approaches exploit the high-fidelity image data in a limited way. In this paper we extract low-level features to characterize the local neighborhood patterns. We exploit the unlabeled feature measurements in a novel way to learn a set of basis functions to derive new features. The derived sparse feature representation obtained by encoding the measured features in terms of the learned basis function set yields superior classification performance. We applied our technique on two challenging image datasets: ORNL dataset representing one-meter spatial resolution satellite imagery representing five land-use categories and, UCMERCED dataset consisting of 21 different categories representing sub-meter resolution overhead imagery. Our results are highly promising and, in the case of UCMERCED dataset we outperform the best results obtained for this dataset. We show that our feature extraction and learning methods are highly effective in developing a detection system that can be used to automatically scan large-scale high-resolution satellite imagery for detecting large-facility.

Basis function↗

Evaluating causal‐based feature selection for fuel property prediction models

Abstract In‐silico screening of novel biofuel molecules based on chemical and fuel properties is a critical first step in the biofuel evaluation process due to the significant volumes of samples required for experimental testing, the destructive nature of engine tests, and the costs associated with bench‐scale synthesis of novel fuels. Predictive models are limited by training sets of few existing measurements, often containing similar classes of molecules that represent just a subset of the potential molecular fuel space. Software tools can be used to generate every possible molecular descriptor for use as input features, but most of these features are largely irrelevant and training models on datasets with higher dimensionality than size tends to yield poor predictive performance. Feature selection has been shown to improve machine learning models, but correlation‐based feature selection fails to provide scientific insight into the underlying mechanisms that determine structure–property relationships. The implementation of causal discovery in feature selection could potentially inform the biofuel design process while also improving model prediction accuracy and robustness to new data. In this study, we investigate the benefits causal‐based feature selection might have on both model performance and identification of key molecular substructures. We found that causal‐based feature selection performed on par with alternative filtration methods, and that a structural causal model provides valuable scientific insights into the relationships between molecular substructures and fuel properties.

Nguyen, Bernard↗

Prognostic Value and Reproducibility of Pretreatment CT Texture Features in Stage III Non-Small Cell Lung Cancer

Purpose: To determine whether pretreatment CT texture features can improve patient risk stratification beyond conventional prognostic factors (CPFs) in stage III non-small cell lung cancer (NSCLC). Methods and Materials: We retrospectively reviewed 91 cases with stage III NSCLC treated with definitive chemoradiation therapy. All patients underwent pretreatment diagnostic contrast enhanced computed tomography (CE-CT) followed by 4-dimensional CT (4D-CT) for treatment simulation. We used the average-CT and expiratory (T50-CT) images from the 4D-CT along with the CE-CT for texture extraction. Histogram, gradient, co-occurrence, gray tone difference, and filtration-based techniques were used for texture feature extraction. Penalized Cox regression implementing cross-validation was used for covariate selection and modeling. Models incorporating texture features from the 33 image types and CPFs were compared to those with models incorporating CPFs alone for overall survival (OS), local-regional control (LRC), and freedom from distant metastases (FFDM). Predictive Kaplan-Meier curves were generated using leave-one-out cross-validation. Patients were stratified based on whether their predicted outcome was above or below the median. Reproducibility of texture features was evaluated using test-retest scans from independent patients and quantified using concordance correlation coefficients (CCC). We compared models incorporating the reproducibility seen on test-retest scans to our original models and determined the classification reproducibility. Results: Models incorporating both texture features and CPFs demonstrated a significant improvement in risk stratification compared to models using CPFs alone for OS (P=.046), LRC (P=.01), and FFDM (P=.005). The average CCCs were 0.89, 0.91, and 0.67 for texture features extracted from the average-CT, T50-CT, and CE-CT, respectively. Incorporating reproducibility within our models yielded 80.4% (±3.7% SD), 78.3% (±4.0% SD), and 78.8% (±3.9% SD) classification reproducibility in terms of OS, LRC, and FFDM, respectively. Conclusions: Pretreatment tumor texture may provide prognostic information beyond that obtained from CPFs. Models incorporating feature reproducibility achieved classification rates of ∼80%. External validation would be required to establish texture as a prognostic factor.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Two-Dimensional Energy Histograms as Features for Machine Learning to Predict Adsorption in Diverse Nanoporous Materials

A major obstacle for machine learning (ML) in chemical science is the lack of physically informed feature representations that provide both accurate prediction and easy interpretability of the ML model. In this work, we describe adsorption systems using novel two-dimensional energy histogram (2D-EH) features, which are obtained from the probe-adsorbent energies and energy gradients at grid points located throughout the adsorbent. The 2D-EH features encode both energetic and structural information of the material and lead to highly accurate ML models (coefficient of determination R2 ~ 0.94–0.99) for predicting single-component adsorption capacity in metal–organic frameworks (MOFs). Here, we consider the adsorption of spherical molecules (Kr and Xe), linear alkanes with a wide range of aspect ratios (ethane, propane, n-butane, and n-hexane), and a branched alkane (2,2-dimethylbutane) over a wide range of temperatures and pressures. The interpretable 2D-EH features enable the ML model to learn the basic physics of adsorption in pores from the training data. We show that these MOF-data-trained ML models are transferrable to different families of amorphous nanoporous materials. We also identify several adsorption systems where capillary condensation occurs, and ML predictions are more challenging. Nevertheless, our 2D-EH features still outperform structural features including those derived from persistent homology. The novel 2D-EH features may help accelerate the discovery and design of advanced nanoporous materials using ML for gas storage and separation in the future.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Atomic-Level Features for Kinetic Monte Carlo Models of Complex Chemistry from Molecular Dynamics Simulations

The high computational cost of evaluating atomic interactions recently motivated the development of computationally inexpensive kinetic models, which can be parametrized from MD simulations of complex chemistry of thousands of species or other processes and accelerate the prediction of the chemical evolution by up to four order of magnitude. Such models go beyond the commonly employed potential energy surface fitting methods in that they are aimed purely at describing kinetic effects. So far, such kinetic models utilize molecular descriptions of reactions and have been constrained to only reproduce molecules previously observed in MD simulations. Therefore, these descriptions fail to predict the reactivity of unobserved molecules, for example in the case of large molecules or solids. In this work, we propose a new approach for the extraction of reaction mechanisms and reaction rates from MD simulations, namely the use of atomic-level features. Using the complex chemical network of hydrocarbon pyrolysis as example, it is demonstrated that kinetic models built using atomic features are able to explore chemical reaction pathways never observed in the MD simulations used to parametrize them, a critical feature to describe rare events. Atomic-level features are shown to construct reaction mechanisms and estimate reaction rates of unknown molecular species from elementary atomic events. Through comparisons of the model ability to extrapolate to longer simulation timescales and different chemical compositions than the ones used for parameterization, it is demonstrated that kinetic models employing atomic features retain the same level of accuracy and transferability as the use of features based on molecular species, while being more compact and parametrized with less data. We also find that atomic features can better describe the formation of large molecules enabling the simultaneous description of small molecules and condensed phases.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Systematic search for tidal features around nearby galaxies: I. Enhanced SDSS imaging of the Local Volume

In hierarchical models of galaxy formation, stellar tidal streams are expected around most, if not all, galaxies. Although these features may provide useful diagnostics of the Λ CDM model, their observational properties remain poorly constrained because they are challenging to detect and interpret and have been studied in detail for only a sparse sampling of galaxy population. More quantitative, systematic approaches are required. We advocate statistical analysis of the counts and properties of such features in archival wide-field imaging surveys for a direct comparison against results from numerical simulations. We aim to study systematically the frequency of occurrence and other observational properties of tidal features around nearby galaxies. The sample we construct will act as a foundational dataset for statistical comparison with cosmological models of galaxy formation. Our approach is based on a visual classification of diffuse features around a volume-limited sample of nearby galaxies, using a post-processing of Sloan Digital Syk Survey (SDSS) imaging optimized for the detection of stellar structure with low surface brightness. At a limiting surface brightness of 28 mag arcsec –2 , 14% of the galaxies in our sample exhibit evidence of diffuse features likely to have arisen from minor merging events. Our technique recovers all previously known streams in our sample and yields a number of new candidates. Consistent with previous studies, coherent arc-like features and shells are the most common type of tidal structures found in this study. Here, we conclude that although some detections are ambiguous and could be corroborated or refuted with deeper imaging, our technique provides a reliable foundation for the statistical analysis of diffuse circumgalactic features in wide-area imaging surveys, and for the identification of targets for follow-up studies.

79 ASTRONOMY AND ASTROPHYSICS↗

UNIDENTIFIED FEATURES IN THE ULTRAVIOLET SPECTRUM OF X Per

High-resolution ultraviolet spectra from the Space Telescope Imaging Spectrograph (STIS) were used to search for unidentified interstellar absorption features in the well studied sight line toward X Per (HD 24534). The significance of features detected was determined from Gaussian fits to the data, as well as the features' persistence in multiple observations. Fixed pattern noise characteristics were studied in STIS echelle data to distinguish between interstellar and instrumental features. We report the detection of two unidentified features that stand out from the more common fixed pattern noise features. Both features have depths of >3% of the continuum level making them very likely of interstellar origin. Lastly, we comment on possible carriers, and discuss future prospects for studying these and perhaps other unidentified lines in larger samples of sight lines.

79 ASTRONOMY AND ASTROPHYSICS↗

FAINT TIDAL FEATURES IN GALAXIES WITHIN THE CANADA-FRANCE-HAWAII TELESCOPE LEGACY SURVEY WIDE FIELDS

We present an analysis of the detectability of faint tidal features in galaxies from the wide-field component of the Canada-France-Hawaii Telescope Legacy Survey. Our sample consists of 1781 luminous (M{sub r{sup '}}<-19.3 mag) galaxies in the magnitude range 15.5 mag < r' < 17 mag and in the redshift range 0.04 < z < 0.2. Although we have classified tidal features according to their morphology (e.g., streams, shells, and tails), we do not attempt to interpret them in terms of their physical origin (e.g., major versus minor merger debris). Instead, we provide a catalog that is intended to provide raw material for future investigations which will probe the nature of low surface brightness substructure around galaxies. We find that around 12% of the galaxies in our sample show clear tidal features at the highest confidence level. This fraction rises to about 18% if we include systems with convincing, albeit weaker tidal features, and to 26% if we include systems with more marginal features that may or may not be tidal in origin. These proportions are a strong function of rest-frame color and of stellar mass. Linear features, shells, and fans are much more likely to occur in massive galaxies with stellar masses >10{sup 10.5} M {sub Sun }, and red galaxies are twice as likely to show tidal features than are blue galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

DUST AROUND R CORONAE BOREALIS STARS. II. INFRARED EMISSION FEATURES IN AN H-POOR ENVIRONMENT

Residual Spitzer/Infrared Spectrograph spectra for a sample of 31 R Coronae Borealis (RCB) stars are presented and discussed in terms of narrow emission features superimposed on the quasi-blackbody continuous infrared emission. A broad {approx}6-10 {mu}m dust emission complex is seen in the RCBs showing an extreme H-deficiency. A secondary and much weaker {approx}11.5-15 {mu}m broad emission feature is detected in a few RCBs with the strongest {approx}6-10 {mu}m dust complex. The Spitzer infrared spectra reveal for the first time the structure within the {approx}6-10 {mu}m dust complex, showing the presence of strong C-C stretching modes at {approx}6.3 and 8.1 {mu}m as well as of other dust features at {approx}5.9, 6.9, and 7.3 {mu}m, which are attributable to amorphous carbonaceous solids with little or no hydrogen. The few RCBs with only moderate H-deficiencies display the classical ''unidentified infrared bands (UIRs)'' and mid-infrared features from fullerene-related molecules. In general, the characteristics of the RCB infrared emission features are not correlated with the stellar and circumstellar properties, suggesting that the RCB dust features may not be dependent on the present physical conditions around RCB stars. The only exception seems to be the central wavelength of the 6.3 {mu}m feature, which is blueshifted in those RCBs showing also the UIRs, i.e., the RCBs with the smallest H deficiency.

79 ASTRONOMY AND ASTROPHYSICS↗

ON IRON MONOXIDE NANOPARTICLES AS A CARRIER OF THE MYSTERIOUS 21 μm EMISSION FEATURE IN POST-ASYMPTOTIC GIANT BRANCH STARS

A prominent mysterious emission feature peaking at ∼20.1 μm—historically known as the '21 μm' feature—is seen in over two dozen Galactic and Magellanic Cloud carbon-rich, post-asymptotic giant branch (post-AGB) stars. The nature of its carrier remains unknown since the first detection of the 21 μm feature in 1989. Over a dozen materials have been suggested as possible carrier candidates. However, none of them has been accepted: they either require too much material (compared to what is available in the circumstellar shells around these post-AGB stars), or exhibit additional emission features that are not seen in these 21 μm sources. Recently, iron monoxide (FeO) nanoparticles seem to be a promising carrier candidate as Fe is an abundant element and FeO emits exclusively at ∼21 μm. In this work, using the proto-typical protoplanetary nebula HD 56126 as a test case, we examine FeO nanoparticles as a carrier for the 21 μm feature by modeling their infrared emission, with FeO being stochastically heated by single stellar photons. We find that FeO emits too broad a 21 μm feature to explain that observed and the Fe abundance required to be locked up in FeO exceeds what is available in HD 56126. We therefore conclude that FeO nanoparticles are not likely to be responsible for the 21 μm feature.

79 ASTRONOMY AND ASTROPHYSICS↗

Untangling features in the primordial spectra

We discuss the possible existence of features in both the primordial power spectrum and bispectrum generated during a stage of single field cosmic inflation. We argue that there are two main classes of features: those produced by a sudden time variation of the sound speed of curvature perturbations, and those produced by a sudden change in the expansion rate during inflation. The former are known to be produced by heavy fields, when the inflationary background trajectory in field space undergoes a bend, whereas the latter are known to be produced by features in the inflaton potential encountered as the inflaton field descends its slope. In general, features are expected to be the result of these two sources combined, however, it is possible that one source dominated over the other, resulting in a distinctive pattern that may be observationally tested. We deduce a relation that gives us the shape of features in the bispectrum provided that we know the shape of features in the power spectrum, and show that each one of these two classes of features leaves a particular footprint in the distribution of perturbations that could be uncovered by a joint analysis of the primordial power spectrum and bispectrum.

79 ASTRONOMY AND ASTROPHYSICS↗

Consistency relations for sharp features in the primordial spectra

We study the generation of sharp features in the primordial spectra within the framework of effective field theory of inflation, wherein curvature perturbations are the consequence of the dynamics of a single scalar degree of freedom. We identify two sources in the generation of features: rapid variations of the sound speed c{sub s} (at which curvature fluctuations propagate) and rapid variations of the expansion rate H during inflation. With this in mind, we propose a non-trivial relation linking these two quantities that allows us to study the generation of sharp features in realistic scenarios where features are the result of the simultaneous occurrence of these two sources. This relation depends on a single parameter with a value determined by the particular model (and its numerical input) responsible for the rapidly varying background. As a consequence, we find a one-parameter consistency relation between the shape and size of features in the bispectrum and features in the power spectrum. To substantiate this result, we discuss several examples of models for which this one-parameter relation (between c{sub s} and H) holds, including models in which features in the spectra are both sudden and resonant.

79 ASTRONOMY AND ASTROPHYSICS↗

Efficient Data-Driven Geologic Feature Characterization from Pre-stack Seismic Measurements using Randomized Machine-Learning Algorithm

Conventional seismic techniques for detecting the subsurface geologic features are challenged by limited data coverage, computational inefficiency and subjective human factors. We developed a novel data-driven geological feature characterization approach based on pre-stack seismic measurements. Our characterization method employs an efficient and accurate machine learning method to extract useful subsurface geologic features automatically. Specifically, we use kernel ridge regression to account for the nonlinear relationship between seismic data and geological features. We moreover employ kernel tricks to avoid the explicit nonlinear mapping and infinite dimension of feature space. Yet, the conventional kernel ridge regression can be computationally prohibitive because of the large volume of seismic measurements. We employ a data reduction technique in combination with the conventional kernel ridge regression method to improve the computational efficiency and reduce memory usage. In particular, we utilize a randomized numerical linear algebra technique, named Nyström method, to effectively reduce the dimensionality of the feature space without compromising the information content required for accurate characterization. We provide thorough computational cost analysis to show the efficiency of our new geological feature characterization methods. We validate the performance of our method in characterizing geologic fault zones because faults play an important role in various subsurface applications. Our numerical examples demonstrate that our new characterization method significantly improves the computational efficiency while maintaining comparable accuracy. Interestingly, we demonstrate that our approach yields a speedup ratio on the order of ~102 to ~103 in a multicore computational environment.

58 GEOSCIENCES↗

Feature selection with distance correlation

Choosing which properties of the data to use as input to multivariate decision algorithms—also known as feature selection—is an important step in solving any problem with machine learning. While there is a clear trend towards training sophisticated deep networks on large numbers of relatively unprocessed inputs (so-called automated feature engineering), for many tasks in physics, sets of theoretically well-motivated and well-understood features already exist. Working with such features can bring many benefits, including greater interpretability, reduced training and run time, and enhanced stability and robustness. We develop a new feature selection method based on distance correlation, and demonstrate its effectiveness on the tasks of boosted top- and W -tagging. Using our method to select features from a set of over 7,000 energy flow polynomials, we show that we can match the performance of much deeper architectures, by using only ten features and two orders-of-magnitude fewer model parameters. Published by the American Physical Society 2024

Astronomy & Astrophysics↗

Exploring Robust Features for Improving Adversarial Robustness

While deep neural networks (DNNs) have revolutionized many fields, their fragility to carefully designed adversarial attacks impedes the usage of DNNs in safety-critical applications. In this article, we strive to explore the robust features that are not affected by the adversarial perturbations, that is, invariant to the clean image and its adversarial examples (AEs), to improve the model’s adversarial robustness. Specifically, we propose a feature disentanglement model to segregate the robust features from nonrobust features and domain-specific features. Here, the extensive experiments on five widely used datasets with different attacks demonstrate that robust features obtained from our model improve the model’s adversarial robustness compared to the state-of-the-art approaches. Moreover, the trained domain discriminator is able to identify the domain-specific features from the clean images and AEs almost perfectly. This enables AE detection without incurring additional computational costs. With that, we can also specify different classifiers for clean images and AEs, thereby avoiding any drop in clean image accuracy.

97 MATHEMATICS AND COMPUTING↗