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 1,693 records · Page 94

LANDSAT Applications for the State of Delaware

The first phase of a cooperative demonstration project between the Eastern Regional Remote Sensing Applications Center and the Delaware Department of Natural Resources and Environmental Control is summarized. Separate land cover classifications were performed for three counties in Delaware using multiseasonal LANDSAT data from April 3 and July 20, 1974. The objective of the New Castle County and Kent County classifications was to provide general land cover information with special emphasis on existing farmland and deciduous and coniferous forest, respectively. A detailed study of existing inland and coastal wetland vegetation communities was attempted for Sussex County. Detailed accuracy assessments were conducted for general land cover in New Castle and Kent Counties and for wetland communities in Sussex County. The classifications were combined to provide a statewide Level I land cover map and acreage statistics. Based on these results, the participating state agencies determined that LANDSAT is a viable tool for mapping and monitoring land cover within the state.

Gervin, J. C.↗

Sparse-Data Deep Learning Strategies for Radiographic Non-Destructive Testing

Radiography is an imaging technique used in a variety of applications, such as medical diagnosis, airport security, and nondestructive testing. We present a deep learning system for extracting information from radiographic images. We perform various prediction tasks using our system, including material classification and regression on the dimensions of a given object that is being radiographed. Our system is designed to address the sparse-data issue for radiographic nondestructive testing applications. It uses a radiographic simulation tool for synthetic data augmentation, and it uses transfer learning with a pre-trained convolutional neural network model. Using this system, our preliminary results indicate that the object geometry regression task saw an improvement of 70% in the R-squared value when using a multi-regime model. In addition, we increase the performance of the object material classification tasks by utilizing data from different imaging systems. In particular, using neutron imaging improved the material classification accuracy by 20% when compared to x-ray imaging.

convolutional neural networks↗

Polyphasic taxonomy of the genus Shewanella and description of Shewanella oneidensis sp. nov

The genus Shewanella has been studied since 1931 with regard to a variety of topics of relevance to both applied and environmental microbiology. Recent years have seen the introduction of a large number of new Shewanella-like isolates, necessitating a coordinated review of the genus. In this work, the phylogenetic relationships among known shewanellae were examined using a battery of morphological, physiological, molecular and chemotaxonomic characterizations. This polyphasic taxonomy takes into account all available phenotypic and genotypic data and integrates them into a consensus classification. Based on information generated from this study and obtained from the literature, a scheme for the identification of Shewanella species has been compiled. Key phenotypic characteristics were sulfur reduction and halophilicity. Fatty acid and quinone profiling were used to impart an additional layer of information. Molecular characterizations employing small-subunit 16S rDNA sequences were at the limits of resolution for the differentiation of species in some cases. As a result, DNA-DNA hybridization and sequence analyses of a more rapidly evolving molecule (gyrB gene) were performed. Species-specific PCR probes were designed for the gyrB gene and used for the rapid screening of closely related strains. With this polyphasic approach, in addition to the ten described Shewanella species, two new species, Shewanella oneidensis and 'Shewanella pealeana', were recognized; Shewanella oneidensis sp. nov. is described here for the first time.

NASA Discipline Exobiology↗

Atlas of Archived Vegetation, Land-use and Seasonal Albedo Data Sets

Global digital data bases of natural vegetation and land use were compiled, for use in climate studies, at 1 deg resolution from over 100 published sources. A series of 6 data sets, derived from the original compilations, was prepared and archived on tape at the National Center for Atmospheric Research (NCAR) (Matthews, 1984). The first is a vegetation data set representing natural (pre-agricultural) vegetation based on the UNESCO classification system. The second, derived from the land-use compilation, is a cultivation-intensity data set defining the areal extent of presently-cultivated land in the 1 deg cells. The last four are integrated surface-albedo data sets (January, April, July, October) for snow-free conditions, incorporating natural-vegetation and cultivation characteristics from the vegetation and cultivation-intensity data sets. Each of these data sets covers the entire surface of the earth. They include non-zero data for permanent land only, including continental ice; water, including oceans and lakes, is zero. The present report includes maps, presented by continent, of the complete archived data, with the exception of Antarctica.

Matthews, E.↗

Using ecological zones to increase the detail of Landsat classifications

Changes in classification detail of forest species descriptions were made for Landsat data on 2.2 million acres in northwestern California. Because basic forest canopy structures may exhibit very similar E-M energy reflectance patterns in different environmental regions, classification labels based on Landsat spectral signatures alone become very generalized when mapping large heterogeneous ecological regions. By adding a seven ecological zone stratification, a 167% improvement in classification detail was made over the results achieved without it. The seven zone stratification is a less costly alternative to the inclusion of complex collateral information, such as terrain data and soil type, into the Landsat data base when making inventories of areas greater than 500,000 acres.

Fox, L., III↗

Classification of US hydropower dams by their modes of operation

A key challenge to understanding ecohydrologic responses to dam regulation is the absence of a universally transferable classification framework for how dams operate. In the present paper, we develop a classification system to organize the modes of operation (MOPs) for U.S. hydropower dams and powerplants. To determine the full diversity of MOPs, we mined federal documents, open-access data repositories, and internet sources. W then used CART classification trees to predict MOPs based on physical characteristics, regulation, and project generation. Finally, we evaluated how much variation MOPs explained in sub-daily discharge patterns for stream gages downstream of hydropower dams. After reviewing information for 721 dams and 597 power plants, we developed a 2-tier hierarchical classification based on 1) the storage and control of flows to powerplants, and 2) the presence of a diversion around the natural stream bed. This resulted in nine tier-1 MOPs representing a continuum of operations from strictly peaking, to reregulating, to run-of-river, and two tier-2 MOPs, representing diversion and integral dam-powerhouse configurations. Although MOPs differed in physical characteristics and energy production, classification trees had low accuracies (<62%), which suggested accurate evaluations of MOPs may require individual attention. MOPs and dam storage explained 20% of the variation in downstream subdaily flow characteristics and showed consistent alterations in subdaily flow patterns from reference streams. Lastly, this standardized classification scheme is important for future research including estimating reservoir operations for large-scale hydrologic models and evaluating project economics, environmental impacts, and mitigation.

13 HYDRO ENERGY↗

A Framework for Identifying Building Energy Models of Localized Utility Service Areas Using Smart Meter Data

Bottom-up load modeling of buildings offers a versatile approach to simulating baseline demand and scenarios of future technology evolution and adoption at the individual building level. This capability is essential to understanding how future load shapes may change with the adoption of electric equipment and vehicles, particularly as it relates to grid planning and infrastructure investments. Traditionally, grid planning techniques have used historical load data to predict future load and infrastructure needs. However, with the anticipated rise in adoption of electrification technologies such as heat pumps and electric vehicles, historical data become less reliable predictors of the future. By employing ResStock, a high-fidelity building stock modeling tool, we can fine-tune electrification scenarios and aggregate models to represent varying geographic resolutions of the grid system, while considering the underlying features of homes. This may enable a more accurate and responsive approach to anticipate and plan for the evolving landscape of energy demands. We present a new framework that leverages building stock energy modeling to identify building models that align with the load shapes and housing attributes of buildings with AMI data. This approach applies two model layers: (1) a classification step that identifies the presence of air conditioning, electric heating, and electric water heating, and (2) an optimization routine that identifies building energy models aligning with load profile data from advanced metering infrastructure meters. This report demonstrates one approach to deploying this framework, and presents results for three test cases that use both modeled and AMI data to assess performance. For a test case using AMI data in Fort Collins, Colorado, we observed a median monthly electricity load CV-RMSE of 16.6%, and a top ten daily heating and cooling median absolute percent error of 7.7% and 8.3%, respectively. For each AMI meter, we identify a set of potential energy models so that downstream use-cases can account for uncertainty driven by variability of baseline technologies and occupant behavior, which impact the response to electrification and energy efficiency scenarios. Our results indicate that ResStock has potential as a scalable solution for modeling residential energy demand at local grid resolutions. Its performance depends on location-specific factors, underlying building characteristics, and the level of aggregation, offering a path towards more precise and adaptive distribution grid planning for the evolving energy landscape.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Investigation of SIR-B images for lithologic mapping

It is immediately apparent from the examination of almost any synthetic aperture radar (SAR) data that the radar return is primarily a function of the topographic relief. Yet radar reflectance is dependent on both surface roughness and the dielectric constant of the surface material. These two parameters can in many cases be related to lithologic units. Thus, if the first-order terrain effects due to topographic relief could (in essence) be removed from the radar image, the SAR data might well be used for lithologic discrimination. Such an approach is evaluated. Landsat Thematic Mapper data along with ground truth are used to define training cells that characterize the various lithologic units in the area. By aggregating the radar data for these cells, curves of radar reflectance versus local incidence are estimated. The curves are then used to classify the test area. The results are compared with lithologic classification based upon multi-spectral (visible and infrared) data.

Parr, J. T.↗

Neutral Networks for Subpixel Classification of Multispectral Images

In this work, the implementation and use of AVHRR (Advanced Very High Resolution Radiometer) images for subpixel analysis is studied. The work consists of two parts, the first is making training data, the second is using the training data to design a neural net subpixel analyzer. Most work on subpixel analysis has been done with images with more spectral bands. AVHRR images were chosen because of their easy acquisition, and because the five spectral bands allow investigation into the development of training data. The first step in subpixel analysis is the development of training data. This consists of image to be classified, and the classification of each pixel. In order to do the classification, a high spatial resolution image is typically needed in order to manually create a classified image. It is difficult to have both the image of interest, and a high spatial resolution image of the same area taken at the same time. Thus it was studied whether a subsampled image taken from the image of interest could serve as the training data. Statistical work has been done showing the unusefulness of this approach. In the second part of the analysis, a feedforward neural net was trained and used to classify the AVHRR images. Results of these tests, comparing the neural net with typical statistical based schemes are shown.

Figueroa, Ricardo R.↗

Parallel membership queries on very large scientific data sets using bitmap indexes

© 2019 John Wiley & Sons, Ltd. Many scientific applications produce very large amounts of data as advances in hardware fuel computing and experimental facilities. Managing and analyzing massive quantities of scientific data is challenging as data are often stored in specific formatted files, such as HDF5 and NetCDF, which do not offer appropriate search capabilities. In this research, we investigated a special class of search capability, called membership query, to identify whether queried elements of a set are members of an attribute. Attributes that naturally have classification values appear frequently in scientific domains such as category and object type as well as in daily life such as zip code and occupation. Because classification attribute values are discrete and require random data access, performing a membership query on a large scientific data set creates challenges. We applied bitmap indexing and parallelization to membership queries to overcome these challenges. Bitmap indexing provides high performance not only for low cardinality attributes but also for high cardinality attributes, such as floating-point variables, electric charge, or momentum in a particle physics data set, due to compression algorithms such as Word-Aligned Hybrid. We conducted experiments, in a highly parallelized environment, on data obtained from a particle accelerator model and a synthetic data set.

Yildiz, Beytullah↗

Analysis of the Westland Data Set

The "Westland" set of empirical accelerometer helicopter data with seeded and labeled faults is analyzed with the aim of condition monitoring. The autoregressive (AR) coefficients from a simple linear model encapsulate a great deal of information in a relatively few measurements; and it has also been found that augmentation of these by harmonic and other parameters call improve classification significantly. Several techniques have been explored, among these restricted Coulomb energy (RCE) networks, learning vector quantization (LVQ), Gaussian mixture classifiers and decision trees. A problem with these approaches, and in common with many classification paradigms, is that augmentation of the feature dimension can degrade classification ability. Thus, we also introduce the Bayesian data reduction algorithm (BDRA), which imposes a Dirichlet prior oil training data and is thus able to quantify probability of error in all exact manner, such that features may be discarded or coarsened appropriately.

Wen, Fang↗

Defect detection in atomic-resolution images via unsupervised learning with translational invariance

Abstract Crystallographic defects can now be routinely imaged at atomic resolution with aberration-corrected scanning transmission electron microscopy (STEM) at high speed, with the potential for vast volumes of data to be acquired in relatively short times or through autonomous experiments that can continue over very long periods. Automatic detection and classification of defects in the STEM images are needed in order to handle the data in an efficient way. However, like many other tasks related to object detection and identification in artificial intelligence, it is challenging to detect and identify defects from STEM images. Furthermore, it is difficult to deal with crystal structures that have many atoms and low symmetries. Previous methods used for defect detection and classification were based on supervised learning, which requires human-labeled data. In this work, we develop an approach for defect detection with unsupervised machine learning based on a one-class support vector machine (OCSVM). We introduce two schemes of image segmentation and data preprocessing, both of which involve taking the Patterson function of each segment as inputs. We demonstrate that this method can be applied to various defects, such as point and line defects in 2D materials and twin boundaries in 3D nanocrystals.

36 MATERIALS SCIENCE↗

SU-G-BRC-13: Model Based Classification for Optimal Position Selection for Left-Sided Breast Radiotherapy: Free Breathing, DIBH, Or Prone

Purpose: There are clinical decision challenges to select optimal treatment positions for left-sided breast cancer patients—supine free breathing (FB), supine Deep Inspiration Breath Hold (DIBH) and prone free breathing (prone). Physicians often make the decision based on experiences and trials, which might not always result optimal OAR doses. We herein propose a mathematical model to predict the lowest OAR doses among these three positions, providing a quantitative tool for corresponding clinical decision. Methods: Patients were scanned in FB, DIBH, and prone positions under an IRB approved protocol. Tangential beam plans were generated for each position, and OAR doses were calculated. The position with least OAR doses is defined as the optimal position. The following features were extracted from each scan to build the model: heart, ipsilateral lung, breast volume, in-field heart, ipsilateral lung volume, distance between heart and target, laterality of heart, and dose to heart and ipsilateral lung. Principal Components Analysis (PCA) was applied to remove the co-linearity of the input data and also to lower the data dimensionality. Feature selection, another method to reduce dimensionality, was applied as a comparison. Support Vector Machine (SVM) was then used for classification. Thirtyseven patient data were acquired; up to now, five patient plans were available. K-fold cross validation was used to validate the accuracy of the classifier model with small training size. Results: The classification results and K-fold cross validation demonstrated the model is capable of predicting the optimal position for patients. The accuracy of K-fold cross validations has reached 80%. Compared to PCA, feature selection allows causal features of dose to be determined. This provides more clinical insights. Conclusion: The proposed classification system appeared to be feasible. We are generating plans for the rest of the 37 patient images, and more statistically significant results are to be presented.

60 APPLIED LIFE SCIENCES↗

Builtin vs. auxiliary detection of extrapolation risk.

A key assumption in supervised machine learning is that future data will be similar to historical data. This assumption is often false in real world applications, and as a result, prediction models often return predictions that are extrapolations. We compare four approaches to estimating extrapolation risk for machine learning predictions. Two builtin methods use information available from the classification model to decide if the model would be extrapolating for an input data point. The other two build auxiliary models to supplement the classification model and explicitly model extrapolation risk. Experiments with synthetic and real data sets show that the auxiliary models are more reliable risk detectors. To best safeguard against extrapolating predictions, however, we recommend combining builtin and auxiliary diagnostics.

Munson, Miles Arthur↗

Digital overlay of cartographic information on Landsat MSS data for soil surveys

Cartographic soils data were digitized, spatially registered, and merged with processed Landsat image data. The Landsat Multispectral Scanner Subsystem (MSS) image data were used to generate a thematic map representing different soil surface characteristics and an enhanced image. The thematic map was generated using supervised and unsupervised classification procedures. The enhanced image was generated by performing a linear contrast stretch on data altered by a principal components transformation. Although both procedures yielded images useful for soil unit delineation, image enhancement was determined to be more suitable because it was more expedient and inexpensive. Enhanced images cost $0.06 per hectare, spectral classifications cost $0.08 per hectare. The overlay of cartographic data on Landsat data facilitates comparisons between the various processing methods used for soil unit boundary determination, delineation, and verification. This technique also provides for accurate and expedient spatial referencing for field observations and cartographic correlation.

Imhoff, M. L.↗

Fault Detection Utilizing Convolution Neural Network on Timeseries Synchrophasor Data From Phasor Measurement Units

An end-to-end supervised learning method is proposed for fault detection in the electric grid using Big Data from multiple Phasor Measurement Units (PMUs). The approach consists of preprocessing steps aimed at reducing data noise and dimensionality, followed by utilization of six classification models considered for detecting faults. Three of the models were variants of Convolutional Neural Network (CNN) architectures that consider a single type of measurement (voltage, current or frequency) at all PMUs or all types together also at all PMUs. CNN based models were compared to traditional methods of Logistic Regression (LR), Multi-layer Perceptron (MLP) and Support Vector Machine (SVM). Evaluation was conducted on two-year data measured by PMUs at 37 locations in a large electric grid. Here, the response variable for classification were extracted from the grid-wide outage event log. Experiments show that CNN-based models outperformed traditional methods on one year out-of-sample outage detection over the entire grid.

42 ENGINEERING↗

Individual variability of protein expression in human tissues

Human tissues are known to exhibit inter-individual proteomic variability, but a deeper understanding of the different factors affecting peptide expression is necessary to further apply this knowledge. Our goal was to explore proteomic discrepancies between individuals as well as between healthy and diseased samples, and to test whether heterogeneous data from many different individuals could be used to build a tissue classifier. In order to investigate whether disparate proteomics datasets may be combined to provide meaningful insights, we performed a retrospective analysis of proteomics data from 9 different human tissues. These datasets represent several different sample prep methods, mass spectrometry instruments, and tissue health. Using this data we examined inter-individual and inter-tissue variability in peptide expression, and analyzed the methods required to build accurate tissue classifiers. We also evaluated the limits of tissue classification, and how results may be improved. These tests provided insights about how classification techniques may be applied to situations where less data is available, such as clinical biopsies, laser capture microdissected samples or potentially to classify cells types based on single cell proteomics. Our findings reveal the strong potential for utilizing proteomics data to build robust tissue classifiers, which has many prospective clinical applications.

Kushner, Irena K.↗