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 307 records · Page 17

Automatic land use classification using Skylab S-192 multispectral data

Investigation of the accuracy attainable in automatic land use classification using 13 bands of multispectral data from the Skylab S-192 scanner. Classification to levels containing seven urban classes, five agricultural, and three water classes is shown to be achievable. With 17 classes, a classification accuracy of 72% was obtained. A wide spectral range, including the thermal band, appears to be most useful for distinguishing urban classes. Agricultural and water classes can be separated using spectral bands covering the visible to far IR.

Kirvida, L.↗

Conjugate-Gradient Neural Networks in Classification of Multisource and Very-High-Dimensional Remote Sensing Data

Application of neural networks to classification of remote sensing data is discussed. Conventional two-layer backpropagation is found to give good results in classification of remote sensing data but is not efficient in training. A more efficient variant, based on conjugate-gradient optimization, is used for classification of multisource remote sensing and geographic data and very-high-dimensional data. The conjugate-gradient neural networks give excellent performance in classification of multisource data, but do not compare as well with statistical methods in classification of very-high-dimentional data.

Benediktsson, J. A.↗

Visualization and Classification of Power System Frequency Data Streams

Two challenges in the realization of the smart grid technology are the ability to visualize the deluge of expected data streams for global situational awareness and the ability to detect disruptive and classify events from spatially-distributed high-speed power system frequency measurements while minimizing false alarms and eliminating missed detection. This paper presents an interactive visualization model for high speed power system frequency data streams that presents both local and global views of the data streams for decision making process. It also presents a K-Median for clustering and identifying disruptive events in spatially-distributed data streams. The results from experimental evaluation on a variety of datasets show that K-Median achieve better performance and empowers analysts with the ability to make sense of a deluge of frequency measurements in a real-time situation.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Power System Feature-Based Event Classification by Means of Multiple PMU Data

Abstract—Phasor Measurement Units (PMUs) provide time synchronized measurements across the power grid, enabling data driven event detection and classification for enhanced system monitoring and situational awareness. However, variations in event duration, spatial extent, and severity, along with coincident events, pose challenges for conventional classification models that require fixed-size inputs. This paper presents a feature-based framework that aggregates diverse attributes from all available PMUs for each event into a fixed-length vector, facilitating the application of standard machine learning classifiers, including Random Forest, XGBoost, and Multilayer Perceptron. A probabilistic post-processing scheme is further introduced to enable multi-label classification in the presence of overlapping events. Experiments using real-world PMU data demonstrate that the Random Forest model achieves 95% accuracy, while the proposed post-processing method yields an additional 3% improvement.

Nematirad, Reza↗

Data-Driven Clustering and Classification of Outage Patterns with Insights into their Links to Extreme Events

At a global level extreme events have increased in both scale and impact. These events have the potential to affect the electrical grid infrastructure and cause a wide range of outages, which can lead to a disruption in daily patterns, cost millions of dollars and also the loss of life. Currently, to track these outage events there have been various approaches developed ranging from regional to national level quantifications for what defines an outage. However, this variation in methods can potentially lead to subjective decision-making and a lack of proper management in relation to the event. While previous work has made strides in determining spatio-temporal patterns, minimal attention has been given to the type and number of outages an area may be exposed to. The differences in incurred cost and the overall severity of an event between a transformer box malfunction and a hurricane are drastic, and by finding historical signals, we can allow for more efficient management, potentially saving lives and millions of dollars. Here, we leverage unsupervised machine learning techniques to delineate outage patterns among 22 counties within the United States and find that there are clear, segregated clusters (0.93 silhouette) of data which are related by event behavior and underlying cause. This finding will allow for energy stakeholders, policy makers, and researchers to gain a deeper understanding of the extent and severity of historic events and to better prepare for electrical grid infrastructure planning and management.

Koob, Benjamin [ORNL]↗

Automated Classification of Transient Contamination in Stationary Acoustic Data

An automated procedure for the classification of transient contamination of stationary acoustic data is proposed and analyzed. The procedure requires the assumption that the stationary acoustic data of interest can be modeled as a band-limited, Gaussian random process. It also requires that the transient contamination be of higher variance than the acoustic data of interest. When these assumptions are satisfied, it is a blind separation procedure, aside from the initial input specifying how to subdivide the time series of interest. No a priori threshold criterion is required. Simulation results show that for a sufficient number of blocks, the method performs well, as long as the occasional false positive or false negative is acceptable. The effectiveness of the procedure is demonstrated with an application to experimental wind tunnel acoustic test data which are contaminated by hydrodynamic gusts.

Bahr, Christopher J.↗

The effect of spatial, spectral and radiometric factors on classification accuracy using thematic mapper data

An experiment of a factorial design was conducted to test the effects on classification accuracy of land cover types due to the improved spatial, spectral and radiometric characteristics of the Thematic Mapper (TM) in comparison to the Multispectral Scanner (MSS). High altitude aircraft scanner data from the Airborne Thematic Mapper instrument was acquired over central California in August, 1983 and used to simulate Thematic Mapper data as well as all combinations of the three characteristics for eight data sets in all. Results for the training sites (field center pixels) showed better classification accuracies for MSS spatial resolution, TM spectral bands and TM radiometry in order of importance.

Wrigley, R. C.↗

The effect of spatial, spectral and radiometric factors on classification accuracy using Thematic Mapper data

An experiment of a factorial design was conducted to test the effects on classification accuracy of land cover types due to the improved spatial, spectral and radiometric characteristics of the LANDSAT Thematic Mapper (TM) in comparison to the Multispectral Scanner (MSS). High altitude aircraft scanner data from the Airborne Thematic Mapper instrument was used to simulate TM data as well as all combinations of the three characteristics for eight data sets in all. Results for the training sites (field-center pixels) show better classification accuracies for MSS spatial resolution, TM spectral bands and TM radiometry in order of importance.

Wrigley, R. C.↗

AGN POPULATION IN HICKSON COMPACT GROUPS. I. DATA AND NUCLEAR ACTIVITY CLASSIFICATION

We have conducted a new spectroscopic survey to characterize the nature of nuclear activity in Hickson compact group (HCG) galaxies and establish its frequency. We have obtained new intermediate-resolution optical spectroscopy for 200 member galaxies and corrected for underlying stellar population contamination using galaxy templates. Spectra for 11 additional galaxies have been acquired from the ESO and 6dF public archives, and emission-line ratios have been taken from the literature for 59 more galaxies. Here we present the results of our classification of the nuclear activity for 270 member galaxies, which belong to a well-defined sample of 64 HCGs. We found a large fraction of galaxies, 63%, with emission lines. Using standard diagnostic diagrams, 45% of the emission-line galaxies were classified as pure active galactic nuclei (AGNs), 23% as Transition Objects (TOs), and 32% as star-forming nuclei (SFNs). In the HCGs, the AGN activity appears as the most frequent activity type. Adopting the interpretation that in TOs a low-luminosity AGN coexists with circumnuclear star formation, the fraction of galaxies with an AGN could rise to 42% of the whole sample. The low frequency (20%) of SFNs confirms that there is no star formation enhancement in HCGs. After extinction correction, we found a median AGN H{alpha} luminosity of 7.1 x 10{sup 39} erg s{sup -1}, which implies that AGNs in HCG have a characteristically low luminosity. This result added to the fact that there is an almost complete absence of broad-line AGNs in compact groups (CGs) as found by MartInez et al. and corroborated in this study for HCGs, is consistent with very few gas left in these galaxies. In general, therefore, what may characterize the level of activity in CGs is a severe deficiency of gas.

79 ASTRONOMY AND ASTROPHYSICS↗

Haze and sun angle effects on automatic classification of satellite data-simulation and correction

Variations in sun angle and haze level change the spectral signatures collected by multispectral scanners (MSS). This paper describes methods and computer programs that have been developed to simulate the effect of such variations and to correct for them. A basic program, Prediction of the Response of Earth Pointed Sensors (PREPS), is used to calculate the response of the sensor as a function of solar angle, atmospheric haze level, and target reflectance. It is then simply a matter of interpolating these results to simulate changes in haze level or solar angle. In principle, this can be done for any sensor, although at the present time it has been completed for only one - the ERTS-1 MSS.

Potter, J. F.↗