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 199 records · Page 11

Evaluation of several schemes for classification of remotely sensed data

Various numerical analysis schemes for the classification of remotely sensed data are evaluated with respect to their capabilities for crop identification. A per point Gaussian maximum likelihood classifier, per point sum-of-normal-densities classifier, per point linear classifier, per point Gaussian maximum likelihood decision tree classifier and a texture-sensitive per field Gaussian maximum likelihood classifier were applied to seven sets of Landsat MSS data on several crop types and regions. The results of the implementation of the classifiers indicate that, given a representative set of training statistics, the choice of classification algorithm of the differentiation of corn and soybeans from one another and from other crop types made relatively little difference in accuracy, whereas the use of a different training method affected the accuracy significantly. In addition, the linear classifier is found to be the easiest for the analyst to use and to cost least in computer time per classification.

Hixson, M.↗

African land-cover classification using satellite data

Data from the advanced very high resolution radiometer sensor on the National Oceanic and Atmospheric Administration's operational series of meteorological satellites were used to classify land cover and monitor vegetation dynamics for Africa over a 19-month period. There was a correspondence between seasonal variations in the density and extent of green leaf vegetation and the patterns of rainfall associated with the movement of the Intertropical Convergence Zone. Regional variations, such as the 1983 drought in the Sahel of western Africa, were observed. Integration of the weekly satellite data with respect to time for a 12-month period produced a remotely sensed estimate of primary production based upon the density and duration of green leaf biomass. Eight of the 21-day composited data sets covering an 11-month period were used to produce a general land-cover classification that corresponded well with those of existing maps.

Tucker, C. J.↗

Procedure 1 and forestland classification using Landsat data

Procedure 1 (P-1) was developed for the Large Area Crop Inventory Experiment (LACIE) and has been used extensively to develop land-use classification of agricultural areas. The P-1 approach requires that pixels (also called dots) of known identity must be located in the study scene. The entire area is clustered and the spectral classes formed are identified using the dots. The analyst need only locate and identify the dots. The rest of the work is done by the computer. The objective of the reported study was to evaluate the effectiveness of P-1's automated approach in a complex forest-land situation. The study site was located in the eastern half of the San Juan National Forest in southwestern Colorado. The study showed that P-1 performed as well as the Multicluster Blocks approach on the rugged study area.

Nelson, R. F.↗

Land use classification for hydrologic models using interactive machine classification of LANDSAT data

Models designed to simulate the hydrology of urban areas require input parameters describing the land use and degree of imperviousness of the watershed. Unfortunately, the magnitude and spatial distribution of these parameters are rather difficult to estimate when a large watershed is involved. Trade-offs between accuracy of the model parameters and the time or money available for their determination must be made. Because of the necessity of such trade-offs, a study was developed to investigate the use of computer aided analysis of LANDSAT multispectral data in estimating percent of imperviousness and associated land uses needed in urban hydrologic modeling. An interactive computer was used to delineate seven land use classifications in the 342 sq. km. Maryland portion of the Anacostia River Basin from LANDSAT data. These results compared favorably with those of an earlier study which obtained the same information through analysis of aerial photographs having a scale of 1:4800. Approximately 94 man days were required to complete the land use analysis using the aerial photographs while less than three man days were required to accomplish similar tasks using the LANDSAT data.

Thomas J. Jackson↗

PMU-data-driven Event Classification in Power Transmission Grids

This paper presents an event classification in transmission grids. The convolutional neural network (CNN)-based classifier is proposed to capture the temporal similarity of time-synchronized data stream from phasor measurement units (PMUs). The proposed CNN is trained using Bayesian optimization to search for the best hyperparameters. The effectiveness of the proposed event classification is validated through the real-world dataset from the U.S. transmission grids. This dataset includes line outage, transformer outage, frequency, and oscillation events. The validation process also includes different PMU outputs, such as voltage magnitude, phase angle, current magnitude, frequency, and rate of change of frequency (ROCOF). The results show that ROCOF gives the best classification performance compared to other PMU outputs. In addition, it is shown that the classifier trained with a larger dataset has higher accuracy. Moreover, the superiority of the proposed method is validated through comparison with other state-of-the-art classification methods.

Niazazari, Iman↗

Characteristics of playa deposits as seen on SIR-A, Seasat and Landsat coregistered data

A classification technique currently under development has been applied for the qualitative study of a playa located in Northeastern Algeria using coregistered SIR-A, Seasat and Landsat MSS 7 data. This classification which is texture based can be applied to only single band imagery at this time. The first results are very encouraging and reach conclusions similar to those obtained with more time consuming classifiers. The signature of each class is well expressed for the Landsat picture. Misclasifications occur with the SIR-A or Seasat images.

Rebillard, PH.↗

The effect of atmospheric water vapor on automatic classification of ERTS data

Absorption by atmospheric water vapor changes the spectral signatures collected by multispectral scanners if channels are not chosen to avoid the atmospheric water bands. For ERTS (Earth Resources Technology Satellite), the Multispectral Scanner band 7 (MSS 7, .8 to 1.1 micron) is the only band significantly affected. Line-by-line atmospheric absorption calculations showed that this effect can multiply the intensity by factors ranging from .77 to 1.0. If horizontal gradients in atmospheric water exist between training fields and the rest of the scene, errors are introduced in automatic classification of the imagery. The degradation of the classification of corn and soybeans was determined by using actual ERTS data and simulating the absorption effects on the MSS 7 band.

Pitts, D. E.↗

Classification of Distributed Data Using Topic Modeling and Maximum Variation Sampling

From a management perspective, understanding the information that exists on a network and how it is distributed provides a critical advantage. This work explores the use of topic modeling as an approach to automatically determine the classes of information that exist on an organization's network, and then use the resultant topics as centroid vectors for the classification of individual documents in order to understand the distribution of information topics across the enterprise network. The approach is tested using the 20 Newsgroups dataset.

Patton, Robert M↗

The Potential of Using Landsat 7 Data for the Classification of Sea Ice Surface Conditions During Summer

During spring and summer, the Surface of the Arctic sea ice cover undergoes rapid changes that greatly affect the surface albedo and significantly impact the further decay of the sea ice. These changes are primarily the development of a wet snow cover and the development of melt ponds. As melt pond diameters generally do not exceed a couple of meters, the spatial resolutions of sensors like AVHRR and MODIS are too coarse for their identification. Landsat 7, on the other hand, has a spatial resolution of 30 m (15 m for the pan-chromatic band). The different wavelengths (bands) from blue to near-infrared offer the potential to distinguish among different surface conditions. Landsat 7 data for the Baffin Bay region for June 2000 have been analyzed. The analysis shows that different surface conditions, such as wet snow and meltponded areas, have different signatures in the individual Landsat bands. Consistent with in-situ albedo measurements, melt ponds show up as blueish whereas dry and wet ice have a white to gray appearance in the Landsat true-color image. These spectral differences enable the distinction of melt ponds. The melt pond fraction for the scene studied in this paper was 37%.

Markus, Thorsten↗

Deep learning as a tool for neural data analysis: Speech classification and cross-frequency coupling in human sensorimotor cortex

A fundamental challenge in neuroscience is to understand what structure in the world is represented in spatially distributed patterns of neural activity from multiple single-trial measurements. This is often accomplished by learning a simple, linear transformations between neural features and features of the sensory stimuli or motor task. While successful in some early sensory processing areas, linear mappings are unlikely to be ideal tools for elucidating nonlinear, hierarchical representations of higher-order brain areas during complex tasks, such as the production of speech by humans. Here, we apply deep networks to predict produced speech syllables from a dataset of high gamma cortical surface electric potentials recorded from human sensorimotor cortex. We find that deep networks had higher decoding prediction accuracy compared to baseline models. Having established that deep networks extract more task relevant information from neural data sets relative to linear models (i.e., higher predictive accuracy), we next sought to demonstrate their utility as a data analysis tool for neuroscience. We first show that deep network's confusions revealed hierarchical latent structure in the neural data, which recapitulated the underlying articulatory nature of speech motor control. We next broadened the frequency features beyond high-gamma and identified a novel high-gamma-to-beta coupling during speech production. Finally, we used deep networks to compare task-relevant information in different neural frequency bands, and found that the high-gamma band contains the vast majority of information relevant for the speech prediction task, with little-to-no additional contribution from lower-frequency amplitudes. Together, these results demonstrate the utility of deep networks as a data analysis tool for basic and applied neuroscience.

60 APPLIED LIFE SCIENCES↗

Geospatial Method for Computing Supplemental Multi-Decadal U.S. Coastal Land-Use and Land-Cover Classification Products, Using Landsat Data and C-CAP Products

This paper discusses the development and implementation of a geospatial data processing method and multi-decadal Landsat time series for computing general coastal U.S. land-use and land-cover (LULC) classifications and change products consisting of seven classes (water, barren, upland herbaceous, non-woody wetland, woody upland, woody wetland, and urban). Use of this approach extends the observational period of the NOAA-generated Coastal Change and Analysis Program (C-CAP) products by almost two decades, assuming the availability of one cloud free Landsat scene from any season for each targeted year. The Mobile Bay region in Alabama was used as a study area to develop, demonstrate, and validate the method that was applied to derive LULC products for nine dates at approximate five year intervals across a 34-year time span, using single dates of data for each classification in which forests were either leaf-on, leaf-off, or mixed senescent conditions. Classifications were computed and refined using decision rules in conjunction with unsupervised classification of Landsat data and C-CAP value-added products. Each classification's overall accuracy was assessed by comparing stratified random locations to available reference data, including higher spatial resolution satellite and aerial imagery, field survey data, and raw Landsat RGBs. Overall classification accuracies ranged from 83 to 91% with overall Kappa statistics ranging from 0.78 to 0.89. The accuracies are comparable to those from similar, generalized LULC products derived from C-CAP data. The Landsat MSS-based LULC product accuracies are similar to those from Landsat TM or ETM+ data. Accurate classifications were computed for all nine dates, yielding effective results regardless of season. This classification method yielded products that were used to compute LULC change products via additive GIS overlay techniques.

Spruce, J. P.↗

A unified approach to the classification of visual data systems.

Description of an approach to attaining a unified means of characterizing film, TV, and optical data systems. The concept is based on the premise that all of these imaging systems can be described by an equation similar to the ideal imaging system described by Rose (1948). This technique permits the direct comparison of film and TV performance without converting speed, film resolution lines, TV resolution lines, highlight output current and video bandwidth into compatible units, only to find some essential element of the conversion has been omitted from the particular specification in use. Most important, it permits system performance criteria to be based on input and output criteria without extensive manipulation of the elements between input and output.

Black, J.↗

Application of information-retrieval methods to the classification of physical data

Scientific data received from satellites are characterized as a multi-dimensional time series, whose terms are vector functions of a vector of measurement conditions. Information retrieval methods are used to construct lower dimensional samples on the basis of the condition vector, in order to obtain these data and to construct partial relations. The methods are applied to the joint Soviet-French Arkad project.

Mamotko, Z. N.↗