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 613 records · Page 34

Effective use of remote sensing products in litigation

A boiled-down version of major legal principles affecting the admissibility of data and products from remote sensing devices is presented. It is suggested that enhancements or classifications of digital data (from scanning devices or from digitized aerial photography) be proffered as evidence in a fashion similar to the manner in which maps from photogrammetric techniques are introduced as evidence. Every effort should be made to illucidate the processes by which digital data are analytically treated or manipulated. Remote sensing expert witnesses should be practiced in providing concise and clear explanations of both data and methods. Special emphasis should be placed on being prepared to provide a detailed accounting of steps taken to calibrate and verify spectral characteristics with ground truth.

Jaynes, R. A.↗

Skylab S192 data evaluation: Comparisons with ERTS-1 results

The author has identified the following significant results. The Skylab S192 data was evaluated by: (1) comparing the classification results using S192 and ERTS-1 data over the Holt County, Nebraska agricultural study area, and (2) investigating the impact of signal-to-noise ratio on classification accuracies using registered S192 and ERTS-1 data. Results indicate: (1) The classification accuracy obtained on S192 data using its best subset of four bands can be expected to be as high as that on ERTS-1 data. (2) When a subset of four S192 bands that are spectrally similar to the ERTS-1 bands was used for classification, an obvious deterioration in the classification accuracy was observed with respect to the ERTS-1 results. (3) The thermal bands 13 and 14 as well as the near IR bands were found to be relatively important in the classification of agricultural data. Although bands 11 and 12 were highly correlated, both were invariably included in the best subsets of the band sizes, four and beyond, according to the divergence criterion. (4) The differentiation of corn from popcorn was difficult on both S192 and ERTS-1 data acquired at an early summer date. (5) The results on both sets of data indicate that it was relatively easy to differentiate grass from any other class.

Chang, C. Y.↗

Characterization of surficial geologic units on Venus from Pioneer Venus radar data: A progress report

A classification database using the reflectivity (derived from the altimetry data), rms slope, and the first principal component of altimetry and topographic slope is presented. The resultant clustered data is examined qualitatively as well as quantitatively, to establish the statistical integrity of each cluster by use of an interactive, ternary plotting algorithm. This algorithm plots, for a cluster, the position of each of its pixels within a ternary diagram whose apices represent reflectivity, rms slope, and the first principal component. The digital values in these three databases are normalized such that unity is represented by a value of 255 in each database. The frequencies of each plotted point within the ternary diagram are recorded in order to establish the mode of each cluster. The pixels of each cluster are displayed as one separate color; their ternary plot will show not only the interrelations between clusters, but also the presence of any anomalous points within a cluster. Existing lunar and terrestrial analog radar data is used to establish fields within this ternary diagram that are indicative of as many different geologic materials and tectonics settings as possible. The resultant fields are used to determine empirically the geologic significance of the clusters resulting from the cluster analysis.

Davis, P. A.↗

Automated Pneumothorax Diagnosis using Deep Neural Networks

Thoracic ultrasound can provide information leading to rapid diagnosis of pneumothorax with improved accuracy over the standard physical examination and with higher sensitivity than anteroposterior chest radiography. However, the clinical We have Furthermore, remote environments, such as the battlefield or deep-space exploration, may lack expertise for diagnosing developed an automated image interpretation pipeline for the analysis of thoracic ultrasound data and the classification of pneumothorax events to provide decision support in such situations. Our pipeline consists of image preprocessing, data augmentation, and deep learning architectures for medical diagnosis. In this work, we demonstrate that robust, accurate interpretation of chest images and video can be achieved using deep neural networks. A number of novel image processing techniques were employed to achieve this result. Affine transformations were applied for data augmentation. Hyperparameters were optimized for learning rate, dropout regularization, batch size, and epoch iteration by a sequential model-based Bayesian approach. In addition, we utilized pretrained architecturesinterpretation of a patient medical image is highly operator dependent. certain pathologies., applying transfer learning and fine-tuning techniques to fully connected layers. Our pipeline yielded binary classification validation accuracies of 98.3% for M-mode images and 99.8% with B-mode video frames.

US Army collaboration↗

A neural network approach to cloud classification

It is shown that, using high-spatial-resolution data, very high cloud classification accuracies can be obtained with a neural network approach. A texture-based neural network classifier using only single-channel visible Landsat MSS imagery achieves an overall cloud identification accuracy of 93 percent. Cirrus can be distinguished from boundary layer cloudiness with an accuracy of 96 percent, without the use of an infrared channel. Stratocumulus is retrieved with an accuracy of 92 percent, cumulus at 90 percent. The use of the neural network does not improve cirrus classification accuracy. Rather, its main effect is in the improved separation between stratocumulus and cumulus cloudiness. While most cloud classification algorithms rely on linear parametric schemes, the present study is based on a nonlinear, nonparametric four-layer neural network approach. A three-layer neural network architecture, the nonparametric K-nearest neighbor approach, and the linear stepwise discriminant analysis procedure are compared. A significant finding is that significantly higher accuracies are attained with the nonparametric approaches using only 20 percent of the database as training data, compared to 67 percent of the database in the linear approach.

Lee, Jonathan↗

Rotational Equivariance for Object Classification Using xView

With the recent addition of large, curated and labeled data sets to the remote sensing discipline, deep learning models have largely surpassed the performance of classical techniques. These deep models, typically Convolutional Neural Networks, are invariant to translation through the use of successive convolution layers which are themselves equivariant to translation. Further, the combination of multiple convolution and pooling layers means that in practice, the model is also approximately invariant to translation. However, until recently these models could only approach rotational invariance through data augmentation. Here we propose using a new model formulation which achieves rotational equaivariance without data augmentation for overhead imagery classification. We utilize the popular xView data set to compare the rotational equivariance formalization against a regular CNN and CNN with rotational data augmentation for the task of image classification.

Bynum, Lucius EJ↗

Block Island Seafloor Sediment and Geological Data

This dataset contains seafloor sediment composition and geological data for the Block Island region, including sediment sample locations, grain size distributions, and seafloor substrate classifications from multiple data sources.

17 WIND ENERGY↗

Bayes classification of terrain cover using normalized polarimetric data

The normalized polarimetric classifier (NPC) which uses only the relative magnitudes and phases of the polarimetric data is proposed for discrimination of terrain elements. The probability density functions (PDFs) of polarimetric data are assumed to have a complex Gaussian distribution, and the marginal PDF of the normalized polarimetric data is derived by adopting the Euclidean norm as the normalization function. The general form of the distance measure for the NPC is also obtained. It is demonstrated that for polarimetric data with an arbitrary PDF, the distance measure of NPC will be independent of the normalization function selected even when the classifier is mistrained. A complex Gaussian distribution is assumed for the polarimetric data consisting of grass and tree regions. The probability of error for the NPC is compared with those of several other single-feature classifiers. The classification error of NPCs is shown to be independent of the normalization function.

Yueh, H. A.↗

Analyzing Multidimensional Image Data

Six computer programs perform histogram cluster analysis. Histogram Cluster Analysis Procedure (HICAP) developed to perform unsupervised classification of multidimensional image data. Clustering approach used in HICAP based on algorithm which uses multidimensional histogram to perform unsupervised classification of four-dimensional Landsat multispectral-scanner data. HICAP generalizes this procedure to process up to 32-bit data with arbitrary number of dimensions. Also incorporates efficiency improvements so classification requires less computation than original algorithm. Computational savings afforded by HICAP increase with number of dimensions in data. HICAP programs written in FORTRAN 77 for batch or interactive execution.

Wharton, S. W.↗

Implementing Legacy-C Algorithms in FPGA Co-Processors for Performance Accelerated Smart Payloads

Accurate, on-board classification of instrument data is used to increase science return by autonomously identifying regions of interest for priority transmission or generating summary products to conserve transmission bandwidth. Due to on-board processing constraints, such classification has been limited to using the simplest functions on a small subset of the full instrument data. FPGA co-processor designs for SVM1 classifiers will lead to significant improvement in on-board classification capability and accuracy.

co-processor↗

Land cover classification for Puget Sound, 1974-1979

Digital analysis of LANDSAT data for land cover classification projects in the Puget Sound region is surveyed. Two early rural and urban land use classifications and their application are described. After acquisition of VICAR/IBIs software, another land use classification of the area was performed, and is described in more detail. Future applications are considered.

Eby, J. R.↗

Previous experience in manned space flight: A survey of human factors lessons learned

Previous experience in manned space flight programs can be used to compile a data base of human factors lessons learned for the purpose of developing aids in the future design of inhabited spacecraft. The objectives are to gather information available from relevant sources, to develop a taxonomy of human factors data, and to produce a data base that can be used in the future for those people involved in the design of manned spacecraft operations. A study is currently underway at the Johnson Space Center with the objective of compiling, classifying, and summarizing relevant human factors data bearing on the lessons learned from previous manned space flights. The research reported defines sources of data, methods for collection, and proposes a classification for human factors data that may be a model for other human factors disciplines.

Chandlee, George O.↗

Advanced Analytics and Big Earth Data

NASA's Earth Science Data Systems process, archive and distribute petabytes of Earth Observation data to a variety of end users. These end users will face dramatically increased data size in the near future, bringing about new challenges and opportunities in analyzing those data. One area of particular ferment currently is Machine Learning. Many Machine Learning methods are black boxes, limiting direct insight into the data's properties. However, they can be used for a variety of data enhancement purposes, such as parameter retrieval, data fusion and image classification and segmentation. The Earth Observing System Data and Information System is also evolving to host large data volumes in the cloud, enabling data proximal analysis. As part of this effort, an Analytics framework is being developed to support and enhance user analysis of the data. By using standards based services in the framework, diverse user communities can be served, while also allowing inter-system collaboration in the analysis process.

Cloud Computing↗

Performance of two texture-based classifiers of cloud fields using spatially averaged Landsat data

Using the gray-level difference vector approach, classification accuracies with 1/8-km spatial-resolution data are similar to those obtained using the full spatial-resolution features. Hence no advantage is to be gained in cloud classification accuracies by using even higher spatial resolutions obtained from Landsat TM or SPOT imagery. The optimum spatial resolution is 1/4 km. However, significant improvement in cloud-classification accuracy compared to that available from the 1-km resolution of AVHRR and GOES imagery is obtained using 1/2-km-resolution data. Cirrus-classification accuracy is especially compromised as spatial resolution is degraded. However, texture measures defined at the combination of pixel separations d = 1,4 improve classification accuracies by several percent, even for 1-km spatial-resolution data. Cirrus-classification accuracy is significantly improved by the use of multiple distance features.

Sengupta, S. K.↗

Photovoltaic System Health-State Architecture for Data-Driven Failure Detection

The timely detection of photovoltaic (PV) system failures is important for maintaining optimal performance and lifetime reliability. A main challenge remains the lack of a unified health-state architecture for the uninterrupted monitoring and predictive performance of PV systems. To this end, existing failure detection models are strongly dependent on the availability and quality of site-specific historic data. The scope of this work is to address these fundamental challenges by presenting a health-state architecture for advanced PV system monitoring. The proposed architecture comprises of a machine learning model for PV performance modeling and accurate failure diagnosis. The predictive model is optimally trained on low amounts of on-site data using minimal features and coupled to functional routines for data quality verification, whereas the classifier is trained under an enhanced supervised learning regime. The results demonstrated high accuracies for the implemented predictive model, exhibiting normalized root mean square errors lower than 3.40% even when trained with low data shares. The classification results provided evidence that fault conditions can be detected with a sensitivity of 83.91% for synthetic power-loss events (power reduction of 5%) and of 97.99% for field-emulated failures in the test-bench PV system. Finally, this work provides insights on how to construct an accurate PV system with predictive and classification models for the timely detection of faults and uninterrupted monitoring of PV systems, regardless of historic data availability and quality. Such guidelines and insights on the development of accurate health-state architectures for PV plants can have positive implications in operation and maintenance and monitoring strategies, thus improving the system’s performance.

photovoltaics↗

Analysis of data acquired by Shuttle Imaging Radar SIR-A and Landsat Thematic Mapper over Baldwin County, Alabama

Seasonally compatible data collected by SIR-A and by Landsat 4 TM over the lower coastal plain in Alabama were coregistered, forming a SIR-A/TM multichannel data set with 30 m x 30 m pixel size. Spectral signature plots and histogram analysis of the data were used to observe data characteristics. Radar returns from pine forest classes correlated highly with the tree ages, suggesting the potential utility of microwave remote sensing for forest biomass estimation. As compared with the TM-only data set, the use of SIR-A/TM data set improved classification accuracy of the seven land cover types studied. In addition, the SIR-A/TM classified data support previous finding by Engheta and Elachi (1982) that microwave data appear to be correlated with differing bottomland hardwood forest vegetation as associated with varying water regimens (i.e., wet versus dry).

Wu, S.-T.↗

Pattern recognition of satellite cloud imagery for improved weather prediction

The major accomplishment was the successful development of a method for extracting time derivative information from geostationary meteorological satellite imagery. This research is a proof-of-concept study which demonstrates the feasibility of using pattern recognition techniques and a statistical cloud classification method to estimate time rate of change of large-scale meteorological fields from remote sensing data. The cloud classification methodology is based on typical shape function analysis of parameter sets characterizing the cloud fields. The three specific technical objectives, all of which were successfully achieved, are as follows: develop and test a cloud classification technique based on pattern recognition methods, suitable for the analysis of visible and infrared geostationary satellite VISSR imagery; develop and test a methodology for intercomparing successive images using the cloud classification technique, so as to obtain estimates of the time rate of change of meteorological fields; and implement this technique in a testbed system incorporating an interactive graphics terminal to determine the feasibility of extracting time derivative information suitable for comparison with numerical weather prediction products.

Gautier, Catherine↗

Self-Taught Anomaly Detection With Hybrid Unsupervised/Supervised Machine Learning in Optical Networks

Here this paper proposes a self-taught anomaly detection framework for optical networks. The proposed framework makes use of a hybrid unsupervised and supervised machine learning scheme. First, it employs an unsupervised data clustering module (DCM) to analyze the patterns of monitoring data. The DCM enables a self-learning capability that eliminates the requirement of prior knowledge of abnormal network behaviors and therefore can potentially detect unforeseen anomalies. Second, we introduce a self-taught mechanism that transfers the patterns learned by the DCM to a supervised data regression and classification module (DRCM). The DRCM, whose complexity is mainly related to the scale of the applied supervised learning model, can potentially facilitate more scalable and time-efficient online anomaly detection by avoiding excessively traversing the original dataset. We designed the DCM and DRCM based on the density-based clustering algorithm and the deep neural network structure, respectively. Evaluations with experimental data from two use cases (i.e., single-point detection and end-to-end detection) demonstrate that up to 99% anomaly detection accuracy can be achieved with a false positive rate below 1%.

42 ENGINEERING↗