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 757 records · Page 42
All sourcesNASA NTRSDOE OSTI
NASA NTRS2024

GRB Progenitor Classification from Gamma-Ray Burst Prompt and Afterglow Observations

Using an established classification technique, we leverage standard observations and analyses to predict the progenitors of gamma-ray bursts (GRBs). This technique, utilizing support vector machine (SVM) statistics, provides a more nuanced prediction than the previous two-component Gaussian mixture in duration of the prompt gamma-ray emission. Based on further covariance testing from Fermi-GBM, Swift-BAT, and Swift-XRT data, we find that our classification based only on prompt emission properties gives perspective on the recent evidence that mergers and collapsars exist in both “long” and “short” GRB populations.

P Nuessle↗
NASA NTRS1981

Evaluation of large area crop estimation techniques using LANDSAT and ground-derived data

The results of the Domestic Crops and Land Cover Classification and Clustering study on large area crop estimation using LANDSAT and ground truth data are reported. The current crop area estimation approach of the Economics and Statistics Service of the U.S. Department of Agriculture was evaluated in terms of the factors that are likely to influence the bias and variance of the estimator. Also, alternative procedures involving replacements for the clustering algorithm, the classifier, or the regression model used in the original U.S. Department of Agriculture procedures were investigated.

Amis, M. L.↗
NASA NTRS1987

Discrimination and supervised classification of volcanic flows of the Puna-Altiplano, Central Andes Mountains using Landsat TM data

Landsat Thematic Mapper (TM) images of portions of the Central Andean Puna-Altiplano volcanic belt have been tested for the feasibility of discriminating individual volcanic flows using supervised classifications. This technique distinguishes volcanic rock classes as well as individual phases (i.e., relative age groups) within each class. The spectral signature of a volcanic rock class appears to depend on original texture and composition and on the degree of erosion, weathering, and chemical alteration. Basalts and basaltic andesite stand out as a clearly distinguishable class. The age dependent degree of weathering of these generally dark volcanic rocks can be correlated with reflectance: older rocks have a higher reflectance. On the basis of this relationship, basaltaic lava flows can be separated into several subclasses. These individual subclasses would correspond to mappable geologic units on the ground at a reconnaissance scale. The supervised classification maps are therefore useful for establishing a general stratigraphic framework for later detailed surface mapping of volcanic sequences.

Mcbride, J. H.↗
NASA NTRS2009

Fast and Adaptive Lossless On-Board Hyperspectral Data Compression System for Space Applications

Efficient on-board lossless hyperspectral data compression reduces the data volume necessary to meet NASA and DoD limited downlink capabilities. The techniques also improves signature extraction, object recognition and feature classification capabilities by providing exact reconstructed data on constrained downlink resources. At JPL a novel, adaptive and predictive technique for lossless compression of hyperspectral data was recently developed. This technique uses an adaptive filtering method and achieves a combination of low complexity and compression effectiveness that far exceeds state-of-the-art techniques currently in use. The JPL-developed 'Fast Lossless' algorithm requires no training data or other specific information about the nature of the spectral bands for a fixed instrument dynamic range. It is of low computational complexity and thus well-suited for implementation in hardware, which makes it practical for flight implementations of pushbroom instruments. A prototype of the compressor (and decompressor) of the algorithm is available in software, but this implementation may not meet speed and real-time requirements of some space applications. Hardware acceleration provides performance improvements of 10x-100x vs. the software implementation (about 1M samples/sec on a Pentium IV machine). This paper describes a hardware implementation of the JPL-developed 'Fast Lossless' compression algorithm on a Field Programmable Gate Array (FPGA). The FPGA implementation targets the current state of the art FPGAs (Xilinx Virtex IV and V families) and compresses one sample every clock cycle to provide a fast and practical real-time solution for Space applications.

image compression↗
NASA NTRS1975

Classification of physiography from ERTS imagery

The potential application of optical data processing to ERTS imagery as a means for automatic identification of large-scale ground patterns was investigated. Spatial frequency distribution and orientational information were derived from ERTS-1 imagery of Kansas for each of 80 sample areas, each 37 km in diameter. The application of classification algorithms to this data reveals that a high degree of correlation exists between the physiography of a sample area and its frequency information. Specifically, the band of frequencies between 1.1 and 2.8 cycles/km appear to contain most of the information needed in distinguishing different physiographic regions.

Ulaby, F. T.↗
NASA NTRS1998

Support Vector Machines for Hyperspectral Remote Sensing Classification

The Support Vector Machine provides a new way to design classification algorithms which learn from examples (supervised learning) and generalize when applied to new data. We demonstrate its success on a difficult classification problem from hyperspectral remote sensing, where we obtain performances of 96%, and 87% correct for a 4 class problem, and a 16 class problem respectively. These results are somewhat better than other recent results on the same data. A key feature of this classifier is its ability to use high-dimensional data without the usual recourse to a feature selection step to reduce the dimensionality of the data. For this application, this is important, as hyperspectral data consists of several hundred contiguous spectral channels for each exemplar. We provide an introduction to this new approach, and demonstrate its application to classification of an agriculture scene.

Gualtieri, J. Anthony↗
NASA NTRS1986

New results for geologic units mapping of Utah test sites using Landsat TM data

This paper continues a study on the accuracy of geological mapping using Landsat Thematic Mapper data (Short, 1984). In June 1976, both the White Mountain alteration zone and the Waterpocket Fold sedimentary rock sites in Utah were surveyed by the Bendix 24-band scanner on a NASA NC-130B aircraft. Mid-June 1984 TM data for these two sites have been processed like the 1976 data to test the quality of simulation of TM data. Principal-components (PC) color composite images for White Mountain show close correspondence to the Bendix PC images. At this site carbonate strata are uniquely discriminated in both Bendix and TM composites that use an inverted PC 3 image. Alunite/kaolinite and hematite/limonite alteration zones developed on volcanic flows are also sharply separated, but iron oxide and silicified zones are less so. The accuracy of rock-units mapping at the Waterpocket Fold site by supervised classification of the June TM data is significantly better, reaching 70 percent in the best case, than for January 1983 data for that site.

Short, N. M.↗
NASA NTRS1980

Numerical trials of HISSE

The mathematical description and implementation of the statistical estimation procedure known as the Houston integrated spatial/spectral estimator (HISSE) is discussed. HISSE is based on a normal mixture model and is designed to take advantage of spectral and spatial information of LANDSAT data pixels, utilizing the initial classification and clustering information provided by the AMOEBA algorithm. The HISSE calculates parametric estimates of class proportions which reduce the error inherent in estimates derived from typical classify and count procedures common to nonparametric clustering algorithms. It also singles out spatial groupings of pixels which are most suitable for labeling classes. These calculations are designed to aid the analyst/interpreter in labeling patches with a crop class label. Finally, HISSE's initial performance on an actual LANDSAT agricultural ground truth data set is reported.

Peters, C.↗
NASA NTRS2009

Hardware Implementation of Lossless Adaptive and Scalable Hyperspectral Data Compression for Space

On-board lossless hyperspectral data compression reduces data volume in order to meet NASA and DoD limited downlink capabilities. The technique also improves signature extraction, object recognition and feature classification capabilities by providing exact reconstructed data on constrained downlink resources. At JPL a novel, adaptive and predictive technique for lossless compression of hyperspectral data was recently developed. This technique uses an adaptive filtering method and achieves a combination of low complexity and compression effectiveness that far exceeds state-of-the-art techniques currently in use. The JPL-developed 'Fast Lossless' algorithm requires no training data or other specific information about the nature of the spectral bands for a fixed instrument dynamic range. It is of low computational complexity and thus well-suited for implementation in hardware. A modified form of the algorithm that is better suited for data from pushbroom instruments is generally appropriate for flight implementation. A scalable field programmable gate array (FPGA) hardware implementation was developed. The FPGA implementation achieves a throughput performance of 58 Msamples/sec, which can be increased to over 100 Msamples/sec in a parallel implementation that uses twice the hardware resources This paper describes the hardware implementation of the 'Modified Fast Lossless' compression algorithm on an FPGA. The FPGA implementation targets the current state-of-the-art FPGAs (Xilinx Virtex IV and V families) and compresses one sample every clock cycle to provide a fast and practical real-time solution for space applications.

FPGA implementation↗
NASA NTRS2014

Airborne Demonstration of FPGA Implementation of Fast Lossless Hyperspectral Data Compression System

Efficient on-board lossless hyperspectral data compression reduces data volume in order to meet NASA and DoD limited downlink capabilities. The technique also improves signature extraction, object recognition and feature classification capabilities by providing exact reconstructed data on constrained downlink resources. At JPL a novel, adaptive and predictive technique for lossless compression of hyperspectral data was recently developed. This technique uses an adaptive filtering method and achieves a combination of low complexity and compression effectiveness that far exceeds state-of-the-art techniques currently in use. The JPL-developed 'Fast Lossless' algorithm requires no training data or other specific information about the nature of the spectral bands for a fixed instrument dynamic range. It is of low computational complexity and thus well-suited for implementation in hardware.

data compression↗
NASA NTRS1978

Apollo 17 mare basalt regression and classification studies

Regression and pattern recognition techniques were applied to 16 chemical species in 34 Apollo 17 basalts. The classification scheme of Pratt et al., (1977) was used. Data were absent for 8 MnO, 3 Hf, 3 Tb, and 8 Cr2O3 analyses. Linear regression studies were utilized to predict these and values obtained were added to the original data base. Pattern recognition techniques were then applied to predict classifications for 30 different Apollo 17 rake basalts analyzed by Murali et al., (1977).

Pratt, D. D.↗
NASA NTRS1987

Global petrologic variations on the moon - A ternary-diagram approach

An improved method for global petrologic mapping of the moon is presented, which uses a color-coded ternary diagram whose apexes are assigned the average Fe and Th/Ti ratio (normalized to chondrites) compositions of KREEP/Mg-suite rocks, mare basalts, and ferroan anorthosites. The composition of each pixel within these Apollo orbital gamma ray data bases is used to determine its position within the ternary diagram. The color corresponding to this ternary position is then placed on a classification map at that pixel's position within the orbital data bases. The resultant classification map shows spatial transitions among petrologic units, allows direct determination of the relative proportions of each end-member composition in a pixel, and provides more geologic information than can be obtained by using only the elemental data. The map units correspond remarkably well to previously delineated geologic provinces.

Davis, Philip A.↗
NASA NTRS1981

Digital ultrasonics signal processing: Flaw data post processing use and description

A modular system composed of two sets of tasks which interprets the flaw data and allows compensation of the data due to transducer characteristics is described. The hardware configuration consists of two main units. A DEC LSI-11 processor running under the RT-11 sngle job, version 2C-02 operating system, controls the scanner hardware and the ultrasonic unit. A DEC PDP-11/45 processor also running under the RT-11, version 2C-02, operating system, stores, processes and displays the flaw data. The software developed the Ultrasonics Evaluation System, is divided into two catagories; transducer characterization and flaw classification. Each category is divided further into two functional tasks: a data acquisition and a postprocessor ask. The flaw characterization collects data, compresses its, and writes it to a disk file. The data is then processed by the flaw classification postprocessing task. The use and operation of a flaw data postprocessor is described.

Buel, V. E.↗
NASA NTRS1984

Analysis methods for Thematic Mapper data of urban regions

Studies have indicated the difficulty in deriving a detailed land-use/land-cover classification for heterogeneous metropolitan areas with Landsat MSS and TM data. The major methodological issues of digital analysis which possibly have effected the results of classification are examined. In response to these methodological issues, a multichannel hierarchical clustering algorithm has been developed and tested for a more complete analysis of the data for urban areas.

Wang, S. C.↗
NASA NTRS1977

Variability of wetland reflectance and its effect on automatic categorization of satellite imagery

The author has identified the following significant results. Land cover categorization of data from the same overpass in four test wetland areas was carried out using a four category classification system. The tests indicate that training data based on in situ reflectance measurements and atmospheric correction of LANDSAT data can produce comparable accuracy of categorization to that achieved using more than four wetlands cover categories (salt marsh cordgrass, salt hay, unvegetated, and water tidal flat) produced overall classification accuracies of 85% by conventional and relative radiance training and 81% by use of in situ measurements. Overall mapping accuracies were 76% and 72% respectively.

Klemas, V.↗
NASA NTRS2000

BOREAS TE-18 Biomass Density Image of the SSA

The BOREAS TE-18 team focused its efforts on using remotely sensed data to characterize the successional and disturbance dynamics of the boreal forest for use in carbon modeling. This biomass density image covers almost the entire BOREAS SSA. The pixels for which biomass density is computed include areas, that are in conifer land cover classes only. The biomass density values represent the amount of overstory biomass (i.e., tree biomass only) per unit area. It is derived from a Landsat-5 TM image collected on 02-Sep-1994. The technique that was used to create this image is very similar to the technique that was as used to create the physical classification of the SSA. The data are provided in a binary image file format. The data files are available on a CD-ROM (see document number 20010000884), or from the Oak Ridge National Laboratory (ORNL) Distributed Activity Archive Center (DAAC).

Hall, Forrest G.↗
NASA NTRS1973

An analysis of Milwaukee county land use

The identification and classification of urban and suburban phenomena through analysis of remotely-acquired sensor data can provide information of great potential value to many regional analysts. Such classifications, particularly those using spectral data obtained from satellites such as the first Earth Resources Technology Satellite (ERTS-1) orbited by NASA, allow rapid frequent and accurate general land use inventories that are of value in many types of spatial analyses. In this study, Milwaukee County, Wisconsin was classified into several broad land use categories on the basis of computer analysis of four bands of ERTS spectral data (ERTS Frame Number E1017-16093). Categories identified were: (1) road-central business district, (2) grass (green vegetation), (3) suburban, (4) wooded suburb, (5) heavy industry, (6) inner city, and (7) water. Overall, 90 percent accuracy was attained in classification of these urban land use categories.

Todd, W. J.↗
NASA NTRS1993

Multifrequency observations

Analyses of synthetic aperture radar (SAR) image data were performed at DLR to classify various kinds of vegetation and different terrain types. The data were collected both with the DLR experimental synthetic aperture radar (E-SAR) in X-band, C-band, and L-band and with the NASA/JPL DC-8 SAR in C-band, L-band, and P-band. E-SAR is a single frequency and single polarization system (both parameters can be selected) but several flights were used to collect multifrequency/multipolarization data which were geometrically matched after processing. Classification of different crop types was based on comparison of the backscatter coefficients of calibrated SAR data in different frequency bands and polarizations. The DC-8 STAR collects polarimetric data in different bands simultaneously. Data acquired with the NASA/JPL DC-8 STAR are qualified for scientific investigations by reducing the cross-talk and channel imbalance to a tolerable extent. The data are absolutely calibrated by using reference targets with known backscattering cross-sections. The signatures and polarimetric features of terrain types, such as grassland, concrete, sea, forest (coniferous; deciduous) and urban areas, are extracted and discussed with respect to frequency and incidence angle dependence. A multifrequency polarimetric feature vector was applied for classification. The results of this new approach for separating and classifying different object classes are presented.

Glitz, R.↗