Search NASA⌕ Search

SEARCH · Search NASA

Results for “data classification”

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 1,765 records · Page 98

classLog: Logistic regression for the classification of genetic sequences

Introduction Sequencing and phylogenetic classification have become a common task in human and animal diagnostic laboratories. It is routine to sequence pathogens to identify genetic variations of diagnostic significance and to use these data in realtime genomic contact tracing and surveillance. Under this paradigm, unprecedented volumes of data are generated that require rapid analysis to provide meaningful inference. Methods We present a machine learning logistic regression pipeline that can assign classifications to genetic sequence data. The pipeline implements an intuitive and customizable approach to developing a trained prediction model that runs in linear time complexity, generating accurate output rapidly, even with incomplete data. Our approach was benchmarked against porcine respiratory and reproductive syndrome virus (PRRSv) and swine H1 influenza A virus (IAV) datasets. Trained classifiers were tested against sequences and simulated datasets that artificially degraded sequence quality at 0, 10, 20, 30, and 40%. Results When applied to a poor-quality sequence data, the classifier achieved between >85% to 95% accuracy for the PRRSv and the swine H1 IAV HA dataset and this increased to near perfect accuracy when using the full dataset. The model also identifies amino acid positions used to determine genetic clade identity through a feature selection ranking within the model. These positions can be mapped onto a maximum-likelihood phylogenetic tree, allowing for the inference of clade defining mutations. Discussion Our approach is implemented as a python package with code available at https://github.com/flu-crew/classLog .

Zeller, Michael A.↗

Initial Verification of GEOS-4 Aerosols Using CALIPSO and MODIS: Scene Classification

A-train sensors such as MODIS and MISR provide column aerosol properties, and in the process a means of estimating aerosol type (e.g. smoke vs. dust). Correct classification of aerosol type is important because retrievals are often dependent upon selection of the right aerosol model. In addition, aerosol scene classification helps place the retrieved products in context for comparisons and analysis with aerosol transport models. The recent addition of CALIPSO to the A-train now provides a means of classifying aerosol distribution with altitude. CALIPSO level 1 products include profiles of attenuated backscatter at 532 and 1064 nm, and depolarization at 532 nm. Backscatter intensity, wavelength ratio, and depolarization provide information on the vertical profile of aerosol concentration, size, and shape. Thus similar estimates of aerosol type using MODIS or MISR are possible with CALIPSO, and the combination of data from all sensors provides a means of 3D aerosol scene classification. The NASA Goddard Earth Observing System general circulation model and data assimilation system (GEOS-4) provides global 3D aerosol mass for sulfate, sea salt, dust, and black and organic carbon. A GEOS-4 aerosol scene classification algorithm has been developed to provide estimates of aerosol mixtures along the flight track for NASA's Geoscience Laser Altimeter System (GLAS) satellite lidar. GLAS launched in 2003 and did not have the benefit of depolarization measurements or other sensors from the A-train. Aerosol typing from GLAS data alone was not possible, and the GEOS-4 aerosol classifier has been used to identify aerosol type and improve the retrieval of GLAS products. Here we compare 3D aerosol scene classification using CALIPSO and MODIS with the GEOS-4 aerosol classifier. Dust, smoke, and pollution examples will be discussed in the context of providing an initial verification of the 3D GEOS-4 aerosol products. Prior model verification has only been attempted with surface mass comparisons and column optical depth from AERONET and MODIS.

Welton, Ellsworth J.↗

Application of LANDSAT system for improving methodology for inventory and classification of wetlands

The author has identified the following significant results. A newly developed software system for generating statistics on surface water features was tested using LANDSAT data acquired previous to 1975. This software test provided a satisfactory evaluation of the system and also allowed expansion of data base on prairie water features. The software system recognizes water on the basis of a classification algorithm. This classification is accomplished by level thresholding a single near infrared data channel. After each pixel is classified as water or nonwater, the software system then recognizes ponds or lakes as sets of contiguous pixels or as single isolated pixels in the case of very small ponds. Pixels are considered to be contiguous if they are adjacent between successive scan lines. After delineating each water feature, the software system then assigns the feature a position based upon a geographic grid system and calculates the feature's planimetric area, its perimeter, and a parameter known as the shape factor.

Gilmer, D. S.↗

Study of the utilization of EREP data from the Wabash River Basin

The author has identified the following significant results. Analysis of the digitized SL/4 S190A color IR photography proved very difficult. An area within Allen County, including Ft. Wayne, was studied. Eight segments of the study area were clustered separately and the cluster maps were then compared with the photography and maps available. The training areas for the land use classes were selected from the cluster maps. The separability measures (transformed divergence) of the classes indicated that many of the land use classes were not spectrally separable. The classification results bore this out. Visually the resulting classification map was poor, with 67 percent correct data. These results were of significantly lower quality than those obtained for the summertime SL/2 data near Lake Monroe, Indiana. Low contrast between land use classes during the wintertime and the limited spectral range and resolution are the major causes for the poor performance.

Silva, L. F.↗

Multi-discipline resource inventory of soils, vegetation and geology

The author has identified the following significant results. Computer classification of natural vegetation, in the vicinity of Big Summit Prairie, Crook County, Oregon was carried out using MSS digital data. Impure training sets, representing eleven vegetation types plus water, were selected from within the area to be classified. Close correlations were visually observed between vegetation types mapped from the large scale photographs and the computer classification of the ERTS data (Frame 1021-18151, 13 August 1972).

Simonson, G. H.↗

Data compression techniques for astronomy

The photometric and astrometric accuracy of the Minnesota automated dual-plate scanner is summarized. The relation between magnitude and isophotometric image diameter is linear on Schmidt plates over a large useful range. Results of a study of centroid measurement accuracy and image classification ability of parameters derived from both densitometric and isophotometric scanning techniques are presented. The data analyzed for this purpose are synthesized ideal star and galaxy brightness distributions with appropriate random noise added to simulate plate grain and sky background effects. A preliminary study of image classification using real plates and single isophote data finds that a parameter describing the raggedness of the isophote may be useful in distinguising stars from galaxies.

Landau, R.↗

A hill-sliding strategy for initialization of Gaussian clusters in the multidimensional space

A hill sliding technique was devised to extract Gaussian clusters from the multivariate probability density estimate of sample data for the first step of iterative unsupervised classification. Each cluster was assumed to posses a unimodal normal distribution. A clustering function proposed distinguished elements of a cluster under formation from the rest in the feature space. Initial clusters were extracted one by one according to the hill sliding tactics. A dimensionless cluster compactness parameter was proposed as a universal measure of cluster goodness and used satisfactorily in test runs with LANDSAT multispectral scanner data. The normalized divergence, defined by the cluster divergence divided by the entropy of the entire sample data, was utilized as a general separability measure between clusters. An overall clustering objective function was set forth in terms of cluster covariance matrices, from which the cluster compactness measure could be deduced. Minimal improvement of initial data partitioning was evaluated by this objective function in eliminating scattered sparse data points. The hill sliding clustering technique developed herein has the potential applicability to decomposition any multivariate mixture distribution into a number of unimodal distributions when an appropriate distribution function to the data set is employed.

Park, J. K.↗

The Evaluation of Machine Learning Techniques for Isotope Identification Contextualized by Training and Testing Spectral Similarity

Precise gamma-ray spectral analysis is crucial in high-stakes applications, such as nuclear security. Research efforts toward implementing machine learning (ML) approaches for accurate analysis are limited by the resemblance of the training data to the testing scenarios. The underlying spectral shape of synthetic data may not perfectly reflect measured configurations, and measurement campaigns may be limited by resource constraints. Consequently, ML algorithms for isotope identification must maintain accurate classification performance under domain shifts between the training and testing data. To this end, four different classifiers (Ridge, Random Forest, Extreme Gradient Boosting, and Multilayer Perceptron) were trained on the same dataset and evaluated on twelve other datasets with varying standoff distances, shielding, and background configurations. A tailored statistical approach was introduced to quantify the similarity between the training and testing configurations, which was then related to the predictive performance. Wilcoxon signed-rank tests revealed that the OVR-wrapped XGB significantly outperformed the other algorithms, with confidence levels of 99.0% or above for the 133Ba, 60Co, 137Cs, and 152Eu sources. The findings from this work are significant as they outline techniques to promote the development of robust ML-based approaches for isotope identification.

domain adaptation↗

Summary of along-track data from the earth radiation budget satellite for several representative ocean regions

For several days in January and August 1985, the Earth Radiation Budget Satellite, a component of the Earth Radiation Budget Experiment (ERBE), was operated in an along-track scanning mode. A survey of radiance measurements taken in this mode is given for five ocean regions: the north and south Atlantic, the Arabian Sea, the western Pacific north of the Equator, and part of the Intertropical Convergence Zone. Each overflight contains information about the clear scene and three cloud categories: partly cloudy, mostly cloudy, and overcast. The data presented include the variation of longwave and shortwave radiance in each scene classification as a function of viewing zenity angle during each overflight of one of the five target regions. Several features of interest in the development of anisotropic models are evident, including the azimuthal dependence of shortwave radiance that is an essential feature of shortwave bidirectional models. The data also demonstrate that the scene classification algorithm employed by the ERBE results in scene classifications that are a function of viewing geometry.

Brooks, David R.↗

Unsupervised classification for region of interest in X-ray ptychography

X-ray ptychography offers high-resolution imaging of large areas at a high computational cost due to the large volume of data provided. To address the cost issue, we propose a physics-informed unsupervised classification algorithm that is performed prior to reconstruction and removes data outside the region of interest (RoI) based on the multimodal features present in the diffraction patterns. The preprocessing time for the proposed method is inconsequential in contrast to the resource-intensive reconstruction process, leading to an impressive reduction in the data workload to a mere 20% of the initial dataset. This capability consequently reduces computational time dramatically while preserving reconstruction quality. Through further segmentation of the diffraction patterns, our proposed approach can also detect features that are smaller than beam size and correctly classify them as within the RoI.

97 MATHEMATICS AND COMPUTING↗

Material identification in x-ray microscopy and micro CT using multi-layer, multi-color scintillation detectors

We demonstrate that a dual-layer, dual-color scintillator construct for microscopic CT, originally proposed to increase sensitivity in synchrotron imaging, can also be used to perform material quantification and classification when coupled with polychromatic illumination. We consider two different approaches to data handling: (1) a data-domain material decomposition whose estimation performance can be characterized by the Cramer-Rao lower bound formalism but which requires careful calibration and (2) an image-domain material classification approach that is more robust to calibration errors. The data-domain analysis indicates that useful levels of SNR (>5) could be achieved in one second or less at typical bending magnet fluxes for relatively large amounts of contrast (several mm path length, such as in a fluid flow experiment) and at typical undulator fluxes for small amount of contrast (tens of microns path length, such as an angiography experiment). The tools introduced could of course be used to study and optimize parameters for a wider range of potential applications. The image domain approach was analyzed in terms of its ability to distinguish different elemental stains by characterizing the angle between the lines traced out in a two-dimensional space of effective attenuation coefficient in the front and back layer images. As a result, this approach was implemented at a synchrotron and the results were consistent with simulation predictions.

36 MATERIALS SCIENCE↗

Stream Classification Tool User Manual: For Use in Applications in Hydropower-Related Evironmental Mitigation

Stream classifications are an inventory of different types of streams. Classifications help us explore similarities and differences among different types of streams, make inferences regarding stream ecosystem behavior, and communicate the complexities of ecosystems. We developed a nested, layered, and spatially contiguous stream classification to characterize the biophysical settings of stream reaches within the Eastern United States (~ 900,000 reaches). The classification is composed of five natural characteristics (hydrology, temperature, size, confinement, and substrate) along with several disturbance regime layers, and each was selected because of their relevance to hydropower mitigation. We developed the classification at the stream reach level using the National Hydrography Dataset Plus Version 1 (1:100k scale). The stream classification is useful to environmental mitigation for hydropower dams in multiple ways. First, it creates efficiency in the regulatory process by creating an objective and data-rich means to address meaningful mitigation actions. Secondly, the SCT addresses data gaps as it quickly provides an inventory of hydrology, temperature, morphology, and ecological communities for the immediate project area, but also surrounding streams. This includes identifying potential reference streams as those that are proximate to the hydropower facility and fall within the same class. These streams can potentially be used to identify ideal environmental conditions or identify desired ecological communities. In doing so, the stream provides some context for how streams may function, respond to dam regulation, and an overview of specific mitigation needs. Herein, we describe the methodology in developing each stream classification layer and provide a tutorial to guide applications of the classification (and associated data) in regulatory settings, such as hydropower (re)licensing.

13 HYDRO ENERGY↗

Deep Learning for Vertex Reconstruction of Neutrino-nucleus Interaction Events with Combined Energy and Time Data

We present a deep learning approach for vertex reconstruction of neutrino-nucleus interaction events, a problem in the domain of high energy physics. In this approach, we combine both energy and timing data that are collected in the MIN-ERvA detector to perform classification and regression tasks. We show that the resulting network achieves higher accuracy than previous results while requiring a smaller model size and less training time. In particular, the proposed model outperforms the state-of-the-art by 4.00% on classification accuracy. For the regression task, our model achieves 0.9919 on the coefficient of determination, higher than the previous work (0.96).

Song, Linghao↗

On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars

Despite the utility of neural networks (NNs) for astronomical time-series classification, the proliferation of learning architectures applied to diverse data sets has thus far hampered a direct intercomparison of different approaches. Here we perform the first comprehensive study of variants of NN-based learning and inference for astronomical time series, aiming to provide the community with an overview on relative performance and, hopefully, a set of best-in-class choices for practical implementations. In both supervised and self-supervised contexts, we study the effects of different time-series-compatible layer choices, namely the dilated temporal convolutional neural network (dTCNs), long-short term memory NNs, gated recurrent units and temporal convolutional NNs (tCNNs). Additionally, we also study the efficacy and performance of encoder-decoder (i.e., autoencoder) networks compared to direct classification networks, different pathways to include auxiliary (non-time-series) metadata, and different approaches to incorporate multi-passband data (i.e., multiple time series per source). Performance—applied to a sample of 17,604 variable stars (VSs) from the MAssive Compact Halo Objects (MACHO) survey across 10 imbalanced classes—is measured in training convergence time, classification accuracy, reconstruction error, and generated latent variables. We find that networks with recurrent NNs generally outperform dTCNs and, in many scenarios, yield to similar accuracy as tCNNs. In learning time and memory requirements, convolution-based layers perform better. We conclude by discussing the advantages and limitations of deep architectures for VS classification, with a particular eye toward next-generation surveys such as the Legacy Survey of Space and Time, the Roman Space Telescope, and Zwicky Transient Facility.

79 ASTRONOMY AND ASTROPHYSICS↗

Radioisotope Identification with List-Mode Gamma-Ray Data

This work explores the potential of utilizing temporal data from gamma-ray detectors, known as list-mode data, to enhance radioisotope identification. Traditional identification methods, which rely on full gamma-ray spectrum analysis, often require long dwell times and struggle with spectra containing similarly spaced spectral peaks. We hypothesize that by leveraging the probabilistic nature of nuclear decay and the time-encoded information from decay sequences and interactions with surrounding materials, we can improve classification accuracy over static spectral analysis. This research examines the temporal content of list-mode data through exploratory data analysis via correlation discovery and qualitative distribution analysis. Additionally, we propose a probabilistic classification model that can utilize spectral data, temporal data, or both to determine if the incorporation of temporal information improves radioisotope identification. Our findings suggest that the temporal information present in list-mode gamma-ray data has merit and should be further investigated to develop more robust and optimal methods for utilizing this temporal information in applications requiring radioisotope identification.

List-mode data↗

A COMPREHENSIVE ANALYSIS OF FERMI GAMMA-RAY BURST DATA. III. ENERGY-DEPENDENT T {sub 90} DISTRIBUTIONS OF GBM GRBs AND INSTRUMENTAL SELECTION EFFECT ON DURATION CLASSIFICATION

The durations (T {sub 90}) of 315 gamma-ray bursts (GRBs) detected with Fermi/GBM (8-1000 keV) up to 2011 September are calculated using the Bayesian Block method. We compare the T {sub 90} distributions between this sample and those derived from previous/current GRB missions. We show that the T {sub 90} distribution of this GRB sample is bimodal, with a statistical significance level comparable to those derived from the BeppoSAX/GRBM sample and the Swift/BAT sample, but lower than that derived from the CGRO/BATSE sample. The short-to-long GRB number ratio is also much lower than that derived from the BATSE sample, i.e., 1:6.5 versus 1:3. We measure T {sub 90} in several bands, i.e., 8-15, 15-25, 25-50, 50-100, 100-350, and 350-1000 keV, to investigate the energy-dependence effect of the bimodal T {sub 90} distribution. It is found that the bimodal feature is well observed in the 50-100 and 100-350 keV bands, but is only marginally acceptable in the 25-50 keV and 350-1000 keV bands. The hypothesis of bimodality is confidently rejected in the 8-15 and 15-25 keV bands. The T {sub 90} distributions in these bands are roughly consistent with those observed by missions with similar energy bands. The parameter T {sub 90} as a function of energy follows T-bar {sub 90}{proportional_to}E{sup -0.20{+-}0.02} for long GRBs. Considering the erratic X-ray and optical flares, the duration of a burst would be even longer for most GRBs. Our results, together with the observed extended emission of some short GRBs, indicate that the central engine activity timescale would be much longer than T {sub 90} for both long and short GRBs and the observed bimodal T {sub 90} distribution may be due to an instrumental selection effect.

79 ASTRONOMY AND ASTROPHYSICS↗

A Comparison of Geospatially Modeled Fire Behavior and Potential Application to Fire and Fuels Management for the Savannah River Site

This study evaluates modeled fire behavior for the Savannah River Site in the Atlantic Coastal Plain of the southeastern U.S. using three data sources: FCCS, LANDFIRE, and SWRA. The Fuel Characteristic Classification System (FCCS) was used to build fuelbeds from intensive field sampling of 629 plots. Custom fire behavior fuel models were derived from these fuelbeds. LANDFIRE developed surface fire behavior fuel models and canopy attributes for the U.S. using satellite imagery informed by field data. The Southern Wildfire Risk Assessment (SWRA) developed surface fire behavior fuel models and canopy cover for the southeastern U.S. using satellite imagery.

60 APPLIED LIFE SCIENCES↗

The ERTS-1 investigation (ER-600). Volume 5: ERTS-1 urban land use analysis

The Urban Land Use Team conducted a year's investigation of ERTS-1 MSS data to determine the number of Land Use categories in the Houston, Texas, area. They discovered unusually low classification accuracies occurred when a spectrally complex urban scene was classified with extensive rural areas containing spectrally homogeneous features. Separate computer processing of only data in the urbanized area increased classification accuracies of certain urban land use categories. Even so, accuracies of urban landscape were in the 40-70 percent range compared to 70-90 percent for the land use categories containing more homogeneous features (agriculture, forest, water, etc.) in the nonurban areas.

Erb, R. B.↗