Engineering Papers⌕ Search

SEARCH · Engineering Papers

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 1,261 records · Page 70
All sourcesNASA NTRSDOE OSTI
NASA NTRS1982

Vegetation survey in Amazonia using LANDSAT data

Automatic Image-100 analysis of LANDSAT data was performed using the MAXVER classification algorithm. In the pilot area, four vegetation units were mapped automatically in addition to the areas occupied for agricultural activities. The Image-100 classified results together with a soil map and information from RADAR images, permitted the establishment of the final legend with six classes: semi-deciduous tropical forest; low land evergreen tropical forest; secondary vegetation; tropical forest of humid areas, predominant pastureland and flood plains. Two water types were identified based on their sediments indicating different geological and geomorphological aspects.

Parada, N. D. J.↗
NASA NTRS1981

Low-cost digital image processing on a university mainframe computer

The advantages and limitations of using university mainframe computers in digital image processing instruction are listed. Aspects to be considered when designing software for this purpose include not only two general audience, but also the capabilities of the system regarding the size of the image/subimage, preprocessing and enhancement functions, geometric correction and registration techniques; classification strategy, classification algorithm, multitemporal analysis, and ancilliary data and geographic information systems. The user/software/hardware interaction as well as acquisition and operating costs must also be considered.

Williams, T. H. L.↗
NASA NTRS1983

Integration of environmental and spectral data for sunflower stress determination

Stress in sunflowers was assessed in western and northwestern Minnesota. Weekly ground observations (acquired in 1980 and 1981) were analyzed in concert with large scale aerial photography and concurrent LANDSAT data. Using multidate supervised and unsupervised classification procedures, it was found that all crops grown in association with sunflowers in the study area are spectrally separable from one another. Under conditions of extreme drought, severely stressed plants were differentiable from those not severely stressed, but between-crop separation was not possible. Initial regression analyses to estimate sunflower seed yield showed a sensitivity to environmental stress during the flowering and seed development stages. One of the most important biological factors related to sunflower production in the Red River Valley area was found to be the extent and severity of insect infestations.

Lillesand, T.↗
NASA NTRS1983

Cloud cover typing from environmental satellite imagery. Discriminating cloud structure with Fast Fourier Transforms (FFT)

The use of two dimensional Fast Fourier Transforms (FFTs) subjected to pattern recognition technology for the identification and classification of low altitude stratus cloud structure from Geostationary Operational Environmental Satellite (GOES) imagery was examined. The development of a scene independent pattern recognition methodology, unconstrained by conventional cloud morphological classifications was emphasized. A technique for extracting cloud shape, direction, and size attributes from GOES visual imagery was developed. These attributes were combined with two statistical attributes (cloud mean brightness, cloud standard deviation), and interrogated using unsupervised clustering amd maximum likelihood classification techniques. Results indicate that: (1) the key cloud discrimination attributes are mean brightness, direction, shape, and minimum size; (2) cloud structure can be differentiated at given pixel scales; (3) cloud type may be identifiable at coarser scales; (4) there are positive indications of scene independence which would permit development of a cloud signature bank; (5) edge enhancement of GOES imagery does not appreciably improve cloud classification over the use of raw data; and (6) the GOES imagery must be apodized before generation of FFTs.

Logan, T. L.↗
NASA NTRS1984

Assessment of LANDSAT for rangeland mapping, Rush Valley, Utah

The feasibility of using LANDSAT MSS (multispectral scanner) data to identify and map cover types for rangeland, and to determine comparative condition of the ecotypes was assessed. A supporting objective is to assess the utility of various forms of aerial photography in the process. If rangelands can be efficiently mapped with Landsat data, as supported by appropriate aerial photography and field data, then uniform standards of cover classification and condition may be applied across the rangelands of the state. Further, a foundation may be established for long-term monitoring of range trend, using the same satellite system over time.

Ridd, M. K.↗
NASA NTRS1987

Correlation between aircraft MSS and lidar remotely sensed data on a forested wetland

Inland wetland in a portion of the Savannah River swamp forest were mapped with an overall accuracy of 88.5 percent on April 26,l985 using high resolution aircraft Daedalus AADS-1268 MSS data. In addition, data were acquired using a NASA sensor system flown along two flight lines over the Steel Creek Delta. The data were significantly correlated with in situ tree height measurements. The data were registered to the wetland classification map and correlated. Statistical analyses demonstrated that the laser derived canopy height information was significantly associated with the Steel Creek Delta wetland classes encountered along the transect (an F-value of 58.46 at the 0.0001 level of confidence). The relationship between vegetation height and vegetation type was then used to produce a three-dimensional model of the landscape which can be of value when computing biomass or canopy density in this forested wetland environment.

Jensen, John R.↗
NASA NTRS1991

Near-Earth asteroids: Observer alert network and database analysis

The Planetary Science Institute (PSI) was funded by SERCulpr to develop a communication network to alert observers of newly discovered near-earth asteroids (NEA's). This network is intended to encourage observers to obtain physical observations of NEA's, which are needed in order to characterize and assess the resource potential of these bodies. This network was declared operational in October 1990 via an announcement to the asteroid observing community. The PSI is also supported to develop the Near-Earth Asteroid Database (NEAD), a comprehensive database of physical and dynamical data on NEA's. In the past year, the database was updated on newly discovered NEA's during 1990, and new data on radar observations and dynamical classifications were added.

Davis, Donald R.↗
NASA NTRS2001

Transport of Energetic Ions in the Ring Current During Geomagnetic Storms

In the final year (plus no-cost extentions) of this grant, we have: Used the particle tracing code to perform a systematic study of the expected energy spectra over the full range of local times in the ring current using a variety of electric and magnetic field models. Shown that the Weimer electric field is superior to the Volland-Stern electric field in reproducing the observed energy spectra on the AMPTE CCE spacecraft. Redone our analysis of the pitch angle spectra of energetic ions during storms in the magnetosphere, using a larger data set, and a more reliable classification technique.

Kistler, Lynn M.↗
NASA NTRS2004

Lunar and Planetary Science XXXV: Image Processing and Earth Observations

The titles in this section include: 1) Expansion in Geographic Information Services for PIGWAD; 2) Modernization of the Integrated Software for Imagers and Spectrometers; 3) Science-based Region-of-Interest Image Compression; 4) Topographic Analysis with a Stereo Matching Tool Kit; 5) Central Avra Valley Storage and Recovery Project (CAVSARP) Site, Tucson, Arizona: Floodwater and Soil Moisture Investigations with Extraterrestrial Applications; 6) ASE Floodwater Classifier Development for EO-1 HYPERION Imagery; 7) Autonomous Sciencecraft Experiment (ASE) Operations on EO-1 in 2004; 8) Autonomous Vegetation Cover Scene Classification of EO-1 Hyperion Hyperspectral Data; 9) Long-Term Continental Areal Reduction Produced by Tectonic Processes.

Source record↗
NASA NTRS2005

Martian Impact Craters as Revealed by MGS and Odyssey

A variety of ejecta and interior morphologies were revealed for martian impact craters by Viking imagery. Numerous studies have classified these ejecta and interior morphologies and looked at how these morphologies correlate with crater diameter, latitude, terrain, and elevation [1, 2, 3, 4]. Many of these features, particularly the layered (fluidized) ejecta morphologies and central pits, have been proposed to result when the crater formed in target material containing high concentrations of volatiles. The Catalog of Large Martian Impact Craters was originally derived from the Viking 1:2,000,000 photomosaics and contains information on 42,283 impact craters 5-km diameter distributed across the entire martian surface. The information in this Catalog has been used to study the distributions of craters displaying specific ejecta and interior morphologies in an attempt to understand the environmental conditions which give rise to these features and to estimate the areal and vertical extents of subsurface volatile reservoirs [4, 5]. The Catalog is currently undergoing revision utilizing Mars Global Surveyor (MGS) and Mars Odyssey data [6]. The higher resolution multispectral imagery is resulting in numerous revisions to the original classifications and the addition of new elemental, thermophysical, and topographic data is allowing new insights into the environmental conditions under which these features form. A few of the new results from analysis of data in the revised Catalog are discussed below.

Barlow, N. G.↗
NASA NTRS2009

A New Monte Carlo Filtering Method for the Diagnosis of Mission-Critical Failures

Testing large-scale systems is expensive in terms of both time and money. Running simulations early in the process is a proven method of finding the design faults likely to lead to critical system failures, but determining the exact cause of those errors is still time-consuming and requires access to a limited number of domain experts. It is desirable to find an automated method that explores the large number of combinations and is able to isolate likely fault points. Treatment learning is a subset of minimal contrast-set learning that, rather than classifying data into distinct categories, focuses on finding the unique factors that lead to a particular classification. That is, they find the smallest change to the data that causes the largest change in the class distribution. These treatments, when imposed, are able to identify the settings most likely to cause a mission-critical failure. This research benchmarks two treatment learning methods against standard optimization techniques across three complex systems, including two projects from the Robust Software Engineering (RSE) group within the National Aeronautics and Space Administration (NASA) Ames Research Center. It is shown that these treatment learners are both faster than traditional methods and show demonstrably better results.

Gay, Gregory↗
NASA NTRS2010

Automatically Finding the Control Variables for Complex System Behavior

Testing large-scale systems is expensive in terms of both time and money. Running simulations early in the process is a proven method of finding the design faults likely to lead to critical system failures, but determining the exact cause of those errors is still time-consuming and requires access to a limited number of domain experts. It is desirable to find an automated method that explores the large number of combinations and is able to isolate likely fault points. Treatment learning is a subset of minimal contrast-set learning that, rather than classifying data into distinct categories, focuses on finding the unique factors that lead to a particular classification. That is, they find the smallest change to the data that causes the largest change in the class distribution. These treatments, when imposed, are able to identify the factors most likely to cause a mission-critical failure. The goal of this research is to comparatively assess treatment learning against state-of-the-art numerical optimization techniques. To achieve this, this paper benchmarks the TAR3 and TAR4.1 treatment learners against optimization techniques across three complex systems, including two projects from the Robust Software Engineering (RSE) group within the National Aeronautics and Space Administration (NASA) Ames Research Center. The results clearly show that treatment learning is both faster and more accurate than traditional optimization methods.

Gay, Gregory↗
NASA NTRS2012

Observing Changing Ecological Diversity in the Anthropocene

As the world enters the Anthropocene, the planet's environment is changing rapidly, putting critical ecosystem services at risk. Understanding and forecasting how ecosystems will change over the coming decades requires understanding the sensitivity of species to environmental change. The extant distribution of species and functional groups contains valuable information about the performance of different species in different environments. However, with high rates of environmental change, information inherent in ranges of many species will disappear, since that information exists only under quasi-equilibrium conditions. The information content of distributional data obtained now is greater than data obtained in the future. New remote sensing technologies can map chemical and structural traits of plant canopies and allow inference of trait and in many cases, species ranges. Current satellite remote sensing data can only produce relatively simple classifications, but new techniques have dramatically higher biological information content.

environmental changes↗
DOE OSTI2022

Nuclear Forensics Scanning

Explore the source record for details and available documents.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

A Catalog of Candidate Double and Lensed Quasars from Gaia and WISE Data

Making use of strong correlations between closely separated multiple or double sources and photometric and astrometric metadata in Gaia Early Data Release 3 (EDR3), we generate a catalog of candidate double- and multiply imaged lensed quasars and active galactic nuclei (AGNs), comprising 3140 systems. It includes two partially overlapping parts: a sample of distant (redshifts mostly greater than 1) sources with perturbed data; and systems that have been resolved into separate components by Gaia at separations less than 2''. For the first part, which is roughly one-third of the published catalog, we synthesized 0.617 million redshifts using multiple machine-learning prediction and classification methods, using independent photometric and astrometric data from Gaia EDR3 and the Wide-field Infrared Survey Explorer, with accurate spectroscopic redshifts from the Sloan Digital Sky Survey (SDSS) as a training set. Using these synthetic redshifts, we estimate a 4.9% rate of interlopers with spectroscopic redshifts below 1 in this part of the catalog. Unresolved candidate double and dual AGNs and quasars are selected as sources with a marginally high BP/RP excess factor (phot_bp_rp_excess_factor), which is sensitive to source extent, limiting our search to high-redshift quasars. For the second part of the catalog, additional filters on measured parallax and near-neighbor statistics are applied to diminish the propagation of the remaining stellar contaminants. The estimated rate of the positives (double or multiple sources) is 98%, and the estimated rate of dual (physically related) quasars is greater than 54%. A few dozen serendipitously found objects of interest are discussed in more detail, including known and new lensed images, planetary nebulae, young IR stars of peculiar morphology, and quasars with catastrophic redshift errors in SDSS.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1975

Results from the crop identification technology assessment for remote sensing /CITARS/ project

The CITARS (Crop Identification Technology Assessment for Remote Sensing) task design, objectives, and results are reviewed along with relevant conclusions and recommendations. The principal assessment concern crop identification performance for corn and soybeans in six sites in Illinois and Indiana. Use of quantitative measures of classification performance and statistical evaluations of the results have been important parts of the technology assessment. Relation of crop and sensor characteristics is discussed. Factors affecting crop identification performance are identified as crop maturity and site characteristics, type of single-date automatic data processing procedure used for local recognition, nonlocal recognition with and without processing for extension of recognition signatures, and use of multidate or multitemporal data. In particular, the probability of correct classification of field center pixels is not well correlated and thus is not a reliable indicator of proportion estimation performance.

Bizzell, R. M.↗
DOE OSTI2021

Deep Cellular Recurrent Network for Efficient Analysis of Time-Series Data With Spatial Information

Efficient processing of large-scale time series data is an intricate problem in machine learning. Conventional sensor signal processing pipelines with hand engineered feature extraction often involve huge computational cost with high dimensional data. Deep recurrent neural networks have shown promise in automated feature learning for improved time-series processing. However, generic deep recurrent models grow in scale and depth with increased complexity of the data. This is particularly challenging in presence of high dimensional data with temporal and spatial characteristics. Consequently, this work proposes a novel deep cellular recurrent neural network (DCRNN) architecture to efficiently process complex multi-dimensional time series data with spatial information. Here, the cellular recurrent architecture in the proposed model allows for location-aware synchronous processing of time series data from spatially distributed sensor signal sources. Extensive trainable parameter sharing due to cellularity in the proposed architecture ensures efficiency in the use of recurrent processing units with high-dimensional inputs. This study also investigates the versatility of the proposed DCRNN model for classification of multi-class time series data from different application domains. Consequently, the proposed DCRNN architecture is evaluated using two time-series datasets: a multichannel scalp EEG dataset for seizure detection, and a machine fault detection dataset obtained in-house. The results suggest that the proposed architecture achieves state-of-the-art performance while utilizing substantially less trainable parameters when compared to comparable methods in the literature.

60 APPLIED LIFE SCIENCES↗
NASA NTRS1975

The use of unsupervised clustering as a classifier for LACIE MSS data

The author has identified the following significant results. This classification method appears to give accurate field center results and to give practical, statistically consistent and accurate estimates of crop proportions. The accuracy of this method is attributable to certain qualities of the particular clustering algorithm. These qualities are freedom from assumptions about Gaussian data, and the continual updating of distribution estimates, including updating the number of modes. This method is relatively tolerant of errors in the determination of crop type, as crop identity is used only for identifying clusters, and not for computing signatures.

Pentland, A. P.↗