Search NASA⌕ Search

SEARCH · Search NASA

Results for “Classifier”

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
All sourcesNASA NTRSDOE OSTI
NASA NTRS1994

Multivariate statistical analysis software technologies for astrophysical research involving large data bases

We developed a package to process and analyze the data from the digital version of the Second Palomar Sky Survey. This system, called SKICAT, incorporates the latest in machine learning and expert systems software technology, in order to classify the detected objects objectively and uniformly, and facilitate handling of the enormous data sets from digital sky surveys and other sources. The system provides a powerful, integrated environment for the manipulation and scientific investigation of catalogs from virtually any source. It serves three principal functions: image catalog construction, catalog management, and catalog analysis. Through use of the GID3* Decision Tree artificial induction software, SKICAT automates the process of classifying objects within CCD and digitized plate images. To exploit these catalogs, the system also provides tools to merge them into a large, complete database which may be easily queried and modified when new data or better methods of calibrating or classifying become available. The most innovative feature of SKICAT is the facility it provides to experiment with and apply the latest in machine learning technology to the tasks of catalog construction and analysis. SKICAT provides a unique environment for implementing these tools for any number of future scientific purposes. Initial scientific verification and performance tests have been made using galaxy counts and measurements of galaxy clustering from small subsets of the survey data, and a search for very high redshift quasars. All of the tests were successful, and produced new and interesting scientific results. Attachments to this report give detailed accounts of the technical aspects for multivariate statistical analysis of small and moderate-size data sets, called STATPROG. The package was tested extensively on a number of real scientific applications, and has produced real, published results.

Djorgovski, S. George↗
NASA NTRS1994

Multivariate Statistical Analysis Software Technologies for Astrophysical Research Involving Large Data Bases

We developed a package to process and analyze the data from the digital version of the Second Palomar Sky Survey. This system, called SKICAT, incorporates the latest in machine learning and expert systems software technology, in order to classify the detected objects objectively and uniformly, and facilitate handling of the enormous data sets from digital sky surveys and other sources. The system provides a powerful, integrated environment for the manipulation and scientific investigation of catalogs from virtually any source. It serves three principal functions: image catalog construction, catalog management, and catalog analysis. Through use of the GID3* Decision Tree artificial induction software, SKICAT automates the process of classifying objects within CCD and digitized plate images. To exploit these catalogs, the system also provides tools to merge them into a large, complex database which may be easily queried and modified when new data or better methods of calibrating or classifying become available. The most innovative feature of SKICAT is the facility it provides to experiment with and apply the latest in machine learning technology to the tasks of catalog construction and analysis. SKICAT provides a unique environment for implementing these tools for any number of future scientific purposes. Initial scientific verification and performance tests have been made using galaxy counts and measurements of galaxy clustering from small subsets of the survey data, and a search for very high redshift quasars. All of the tests were successful and produced new and interesting scientific results. Attachments to this report give detailed accounts of the technical aspects of the SKICAT system, and of some of the scientific results achieved to date. We also developed a user-friendly package for multivariate statistical analysis of small and moderate-size data sets, called STATPROG. The package was tested extensively on a number of real scientific applications and has produced real, published results.

Djorgovski, S. G.↗
NASA NTRS1993

Classification of high dimensional multispectral image data

A method for classifying high dimensional remote sensing data is described. The technique uses a radiometric adjustment to allow a human operator to identify and label training pixels by visually comparing the remotely sensed spectra to laboratory reflectance spectra. Training pixels for material without obvious spectral features are identified by traditional means. Features which are effective for discriminating between the classes are then derived from the original radiance data and used to classify the scene. This technique is applied to Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data taken over Cuprite, Nevada in 1992, and the results are compared to an existing geologic map. This technique performed well even with noisy data and the fact that some of the materials in the scene lack absorption features. No adjustment for the atmosphere or other scene variables was made to the data classified. While the experimental results compare favorably with an existing geologic map, the primary purpose of this research was to demonstrate the classification method, as compared to the geology of the Cuprite scene.

Hoffbeck, Joseph P.↗
NASA NTRS1995

Bayes classification of interferometric TOPSAR data

We report the Bayes classification of terrain types at different sites using airborne interferometric synthetic aperture radar (INSAR) data. A Gaussian maximum likelihood classifier was applied on multidimensional observations derived from the SAR intensity, the terrain elevation model, and the magnitude of the interferometric correlation. Training sets for forested, urban, agricultural, or bare areas were obtained either by selecting samples with known ground truth, or by k-means clustering of random sets of samples uniformly distributed across all sites, and subsequent assignments of these clusters using ground truth. The accuracy of the classifier was used to optimize the discriminating efficiency of the set of features that was chosen. The most important features include the SAR intensity, a canopy penetration depth model, and the terrain slope. We demonstrate the classifier's performance across sites using a unique set of training classes for the four main terrain categories. The scenes examined include San Francisco (CA) (predominantly urban and water), Mount Adams (WA) (forested with clear cuts), Pasadena (CA) (urban with mountains), and Antioch Hills (CA) (water, swamps, fields). Issues related to the effects of image calibration and the robustness of the classification to calibration errors are explored. The relative performance of single polarization Interferometric data classification is contrasted against classification schemes based on polarimetric SAR data.

Michel, T. R.↗
NASA NTRS1994

Coccolithophorid blooms in the global ocean

The global distribution pattern of coccolithophrid blooms was mapped in order to ascertain the prevalence of these blooms in the world's oceans and to estimate their worldwide production of CaCO3 and dimethyl sulfide (DMS). Mapping was accomplished by classifying pixels of 5-day global composites of coastal zone color scanner imagery into bloom and nonbloom classes using a supervised, multispectral classification scheme. Surface waters with the spectral signature of coccolithophorid blooms annually covered an average of 1.4 x 10(exp 6) sq km in the world oceans from 1979 to 1985, with the subpolar latitudes accounting for 71% of this surface area. Classified blooms were most extensive in the Subartic North Atlantic. Large expanses of the bloom signal were also detected in the North Pacific, on the Argentine shelf and slope, and in numerous lower latitude marginal seas and shelf regions. The greatest spatial extent of classified blooms in subpolar oceanic regions occurred in the months from summer to early autumn, while those in lower latitude marginal seas occurred in midwinter to early spring. Though the classification scheme was effcient in separating bloom and nonbloom classes during test simulations, and biogeographical literature generally confirms the resulting distribution pattern of blooms in the subpolar regions, the cause of the bloom signal is equivocal in some geographic areas, particularly on shelf regions at lower latitudes. Standing stock estimates suggest that the presumed Emiliania huxleyi blooms act as a significant source of calcite carbon and DMS sulfur on a regional scale. On a global scale, however, the satellite-detected coccolithophorid blooms are estimated to play only a minor role in the annual production of these two compounds and their flux from the surface mixed layer.

Brown, Christopher W.↗
NASA NTRS1994

Feasibility of sea ice typing with synthetic aperture radar (SAR): Merging of Landsat thematic mapper and ERS 1 SAR satellite imagery

Earth Remote-Sensing Satellite (ERS) 1 synthetic aperture radar (SAR) and Landsat thematic mapper (TM) images were acquired for the same area in the Beaufort Sea, April 16 and 18, 1992. The two image pairs were colocated to the same grid (25-m resolution), and a supervised ice type classification was performed on the TM images in order to classify ice free, nilas, gray ice, gray-white ice, thin first-year ice, medium and thick first-year ice, and old ice. Comparison of the collocated SAR pixels showed that ice-free areas can only be classified under calm wind conditions (less than 3 m/s) and for surface winds greater than 10 m/s based on the backscattering coefficient alone. This is true for pack ice regions during the cold months of the year where ice-free areas are spatially limited and where the capillary waves that cause SAR backscatter are dampened by entrained ice crystals. For nilas, two distinct backscatter classes were found at -17 dB and at -10 dB. The higher backscattering coefficient is attributed to the presence of frost flowers on light nilas. Gray and gray-white ice have a backscatter signature similar to first-year ice and therefore cannot be distinguished by SAR alone. First-year and old ice can be clearly separated based on their backscattering coefficient. The performance of the Geophysical Processor System ice classifier was tested against the Landsat derived ice products. It was found that smooth first-year ice and rough first-year ice were not significantly different in the backscatter domain. Ice concentration estimates based on ERS 1 C band SAR showed an error range of 5 to 8% for high ice concentration regions, mainly due to misclassified ice-free and smooth first-year ice areas. This error is expected to increase for areas of lower ice concentration. The combination of C band SAR and TM channels 2, 4, and 6 resulted in ice typing performance with an estimated accuracy of 90% for all seven ice classes.

Steffen, Konrad↗
NASA NTRS1996

Software For Multivariate Bayesian Classification

PHD general-purpose classifier computer program. Uses Bayesian methods to classify vectors of real numbers, based on combination of statistical techniques that include multivariate density estimation, Parzen density kernels, and EM (Expectation Maximization) algorithm. By means of simple graphical interface, user trains classifier to recognize two or more classes of data and then use it to identify new data. Written in ANSI C for Unix systems and optimized for online classification applications. Embedded in another program, or runs by itself using simple graphical-user-interface. Online help files makes program easy to use.

Saul, Ronald↗
NASA NTRS1993

Analyzing High-Dimensional Multispectral Data

In this paper, through a series of specific examples, we illustrate some characteristics encountered in analyzing high- dimensional multispectral data. The increased importance of the second-order statistics in analyzing high-dimensional data is illustrated, as is the shortcoming of classifiers such as the minimum distance classifier which rely on first-order variations alone. We also illustrate how inaccurate estimation or first- and second-order statistics, e.g., from use of training sets which are too small, affects the performance of a classifier. Recognizing the importance of second-order statistics on the one hand, but the increased difficulty in perceiving and comprehending information present in statistics derived from high-dimensional data on the other, we propose a method to aid visualization of high-dimensional statistics using a color coding scheme.

Lee, Chulhee↗
NASA NTRS1996

Feature Selection for Classification of Polar Regions Using a Fuzzy Expert System

Labeling, feature selection, and the choice of classifier are critical elements for classification of scenes and for image understanding. This study examines several methods for feature selection in polar regions, including the list, of a fuzzy logic-based expert system for further refinement of a set of selected features. Six Advanced Very High Resolution Radiometer (AVHRR) Local Area Coverage (LAC) arctic scenes are classified into nine classes: water, snow / ice, ice cloud, land, thin stratus, stratus over water, cumulus over water, textured snow over water, and snow-covered mountains. Sixty-seven spectral and textural features are computed and analyzed by the feature selection algorithms. The divergence, histogram analysis, and discriminant analysis approaches are intercompared for their effectiveness in feature selection. The fuzzy expert system method is used not only to determine the effectiveness of each approach in classifying polar scenes, but also to further reduce the features into a more optimal set. For each selection method,features are ranked from best to worst, and the best half of the features are selected. Then, rules using these selected features are defined. The results of running the fuzzy expert system with these rules show that the divergence method produces the best set features, not only does it produce the highest classification accuracy, but also it has the lowest computation requirements. A reduction of the set of features produced by the divergence method using the fuzzy expert system results in an overall classification accuracy of over 95 %. However, this increase of accuracy has a high computation cost.

Penaloza, Mauel A.↗
NASA NTRS1999

Physical and Chemical Characterization of Particles in the Upper Troposphere and Lower Stratosphere: Microanalysis of Aerosol Impactor Samples

Herein is reported activities to support the characterization of the aerosol in the upper troposphere (UT) and lower stratosphere (LS) collected during the Airborne Southern Hemisphere Ozone Experiment/Measurements for Assessing the Effects of Stratospheric Aircraft (ASHOE/MAESA) missions in 1994. Through a companion proposal, another group was to measure the size distribution of aerosols in the 0.008 to 2 micrometer diameter range and to collect for us impactor samples of particles larger than about 0.02 gm. In the first year, we conducted laboratory studies related to particulate deposition patterns on our collection substrates, and have performed the analysis of many ASHOE/MAESA aerosol samples from 1994 using analytical electron microscopy (AEM). We have been building an "aerosol climatology" with these data that documents the types and relative abundances of particles observed at different latitudes and altitudes. The second year (and non-funded extension periods) saw continued analyses of impactor aerosol samples, including more ASHOE/MAESA samples, some northern hemisphere samples from the NASA Stratospheric Photochemistry Aerosols and Dynamics Expedition (SPADE) program for comparison, and a few aerosol samples from the NASA Stratospheric TRacers of Atmospheric Transport (STRAT) program. A high-resolution field emission microscope was used for the analysis and re-analysis of a number of samples to determine if this instrument was superior in performance to our conventional electron microscope. In addition, some basic laboratory studies were conducted to determine the minimum detectable and analyzable particle size for different types of aerosols. In all, 61 aerosol samples were analyzed, with a total of over 30,000 individual particle analyses. In all analyzed samples, sulfate particles comprised the major aerosol number fraction. It must be stressed that particles composed of more than one species, for example sulfate and organic carbon, were classified according to the major fraction. Thus, many of the particles classified as sulfate may have contained significant mass fractions of carbonaceous or other material. These particles for the most part did not show two physical phases, however. Nonsulfate particles were classified according to the physical and chemical characteristics of each particle, and were grouped into the major nonsulfate particle classes, including C-rich, crustal, metallic, and salts. Our UT and LS sample analyses indicate a maximum for crustal and C-rich particle abundance in the Northern Hemisphere upper troposphere, and a salt particle maximum in the Southern Hemisphere upper troposphere. Metallic particles are clearly more prevalent in the troposphere than in the stratosphere, but interhemispheric differences appear small.

Sheridan, Patrick J.↗
NASA NTRS1998

Statistical Aspects of Intense Hurricanes in the Atlantic Basin during the Past 49 Hurricane Seasons (1950-1998)

Statistical aspects of intense hurricanes (those of category 3 or higher) in the Atlantic basin for the interval of 1950-1998 are investigated with respect to the ENSC, cycle and the postulated 'more versus less' activity modes for intense hurricane activity. This is accomplished in order to evaluate the statistical basis for these specific groupings and to presage a forecast for the number of expected intense hurricanes during the forthcoming 1999 hurricane season. Statistically speaking, because the 1998-1999 La Nina (onset presumed to be September 1998) should persist longer than about a year - the average duration for La Nina events based on the 10 previously occurring La Nina, each having a duration in the range of 7-22 months - it seems likely that the 1999 season will be classified as a 'non-El Nino-related' (NENR) season. If true, then, greater than or equal to 2 intense hurricanes are to be expected. Based on Poisson statistics, the probability of greater than or equal to 2 events is about 77% when the season is classified as NENR, and it is about 87% when the season is classified as NENR and the 'more active' phase is in fashion; likewise, the probability of greater than or equal to 4 events is about 31% and 48%, respectively, for the two cases. Therefore, an above average rate (possibly, as many as 4 +/- 1, or higher) of intense hurricanes forming in the Atlantic basin seems a very distinct possibility during the 1999 season.

Wilson, Robert M.↗
NASA NTRS1999

Statistical Aspects of Intense Hurricanes in the Atlantic Basin during the Past 49 Hurricane Seasons (1950-1998): Implications for the Current Season

Statistical aspects of intense hurricanes (those of category 3 or higher) in the Atlantic basin during the interval of 1950-1998 are investigated in relation to the ENSO cycle and to the postulated 'more versus less' activity modes for intense hurricane activity. Because the 1999 hurricane season likely will be classified as a 'non-El Nino-related' (NENR) season and that the more active mode appears to be in vogue, an above average seasonal rate of greater than or equal to 2 intense hurricanes is to be expected (probably, about 4 +/- 1, or higher). Based on Poisson statistics, when the hurricane season is classified as NENR the probability of greater than or equal to 2 events is about 77%, whereas when the season is classified as NENR and the more active mode is operative it is about 87%. The probability of greater than or equal to 4 events is about 31 % and 48 %, respectively, for these two activity classes.

Wilson, Robert M.↗
NASA NTRS1998

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.↗
NASA NTRS2000

Mapping Land Cover Types in Amazon Basin Using 1km JERS-1 Mosaic

In this paper, the 100 meter JERS-1 Amazon mosaic image was used in a new classifier to generate a I km resolution land cover map. The inputs to the classifier were 1 km resolution mean backscatter and seven first order texture measures derived from the 100 m data by using a 10 x 10 independent sampling window. The classification approach included two interdependent stages: 1) a supervised maximum a posteriori Bayesian approach to classify the mean backscatter image into 5 general land cover categories of forest, savannah, inundated, white sand, and anthropogenic vegetation classes, and 2) a texture measure decision rule approach to further discriminate subcategory classes based on taxonomic information and biomass levels. Fourteen classes were successfully separated at 1 km scale. The results were verified by examining the accuracy of the approach by comparison with the IBGE and the AVHRR 1 km resolution land cover maps.

Saatchi, Sassan S.↗
NASA NTRS2000

Information Theoretic Extraction of EEG Features for Monitoring Subject Attention

The goal of this project was to test the applicability of information theoretic learning (feasibility study) to develop new brain computer interfaces (BCI). The difficulty to BCI comes from several aspects: (1) the effective data collection of signals related to cognition; (2) the preprocessing of these signals to extract the relevant information; (3) the pattern recognition methodology to detect reliably the signals related to cognitive states. We only addressed the two last aspects in this research. We started by evaluating an information theoretic measure of distance (Bhattacharyya distance) for BCI performance with good predictive results. We also compared several features to detect the presence of event related desynchronization (ERD) and synchronization (ERS), and concluded that at least for now the bandpass filtering is the best compromise between simplicity and performance. Finally, we implemented several classifiers for temporal - pattern recognition. We found out that the performance of temporal classifiers is superior to static classifiers but not by much. We conclude by stating that the future of BCI should be found in alternate approaches to sense, collect and process the signals created by populations of neurons. Towards this goal, cross-disciplinary teams of neuroscientists and engineers should be funded to approach BCIs from a much more principled view point.

Principe, Jose C.↗
NASA NTRS1999

Regional Climatology and Surface Radiation Budget

The climatology and surface radiation budget (SRB) of a region are intimately related. This paper presents a brief examination of this relationship. An 8-year surface radiation budget data set has been developed based on satellite measurements. In that data set and in this paper a region is defined as a quasi-square 2.5o in latitude and approximately the same physical distance in longitude. A pilot study by Wilber et al. (1998) showed a variety of behaviors of the annual cycles for selected regions. Selected desert regions form a loop in a specific part of the plot, with large NLW and large NSW. Tropical wet regions form much smaller loops in a different part of the plot, with small NLW and large NSW. For regions selected in high latitude the annual cycles form nearly linear figures in another part of the plot. The question arises as to whether these trajectories are characteristic of the climatology of the region or simply the behavior of the few regions selected from the set of 6596 regions. In order to address this question, it is necessary to classify the climatology of the each region, e.g. as classified by Koeppen (1936) or Trenwarthe and Horne (1980). This paper presents a method of classifying climate of the regions on the basis of the surface radiation behavior such that the results are very similar to the classification of Trenwarthe and Horne. The characteristics of the annual cycle of SRB components can then be investigated further, based on the climate classification of each region.

Wilber, Anne C.↗
NASA NTRS2004

Measures and Models for Estimating and Predicting Cognitive Fatigue

We analyzed EEG and ERPs in a fatiguing mental task and created statistical models for single subjects. Seventeen subjects (4 F, 18-38 y) viewed 4-digit problems (e.g., 3+5-2+7=15) on a computer, solved the problems, and pressed keys to respond (intertrial interval = 1 s). Subjects performed until either they felt exhausted or three hours had elapsed. Re- and post-task measures of mood (Activation Deactivation Adjective Checklist, Visual Analogue Mood Scale) confirmed that fatigue increased and energy decreased over time. We tested response times (RT); amplitudes of ERP components N1, P2, P300, readiness potentials; and amplitudes of frontal theta and parietal alpha rhythms for change as a function of time. For subjects who completed 3 h (n=9) we analyzed 12 15-min blocks. For subjects who completed at least 1.5 h (n=17), we analyzed the first-, middle-, and last 100 error-free trials. Mean RT rose from 6.7 s to 8.5 s over time. We found no changes in the amplitudes of ERP components. In both analyses, amplitudes of frontal theta and parietal alpha rose by 30% or more over time. We used 30-channel EEG frequency spectra to model the effects of time in single subjects using a kernel partial least squares classifier. We classified 3.5s EEG segments as being from the first 100 or the last 100 trials, using random sub-samples of each class. Test set accuracies ranged from 63.9% to 99.6% correct. Only 2 of 17 subjects had mean accuracies lower than 80%. The results suggest that EEG accurately classifies periods of cognitive fatigue in 90% of subjects.

Trejo, Leonard J.↗
NASA NTRS2005

Brain-Computer Interfaces for 1-D and 2-D Cursor Control: Designs Using Volitional Control of the EEG Spectrum or Steady-State Visual Evoked Potentials

We have developed and tested two EEG-based brain-computer interfaces (BCI) for users to control a cursor on a computer display. Our system uses an adaptive algorithm, based on kernel partial least squares classification (KPLS), to associate patterns in multichannel EEG frequency spectra with cursor controls. Our first BCI, Target Practice, is a system for one-dimensional device control, in which participants use biofeedback to learn voluntary control of their EEG spectra. Target Practice uses a KF LS classifier to map power spectra of 30-electrode EEG signals to rightward or leftward position of a moving cursor on a computer display. Three subjects learned to control motion of a cursor on a video display in multiple blocks of 60 trials over periods of up to six weeks. The best subject s average skill in correct selection of the cursor direction grew from 58% to 88% after 13 training sessions. Target Practice also implements online control of two artifact sources: a) removal of ocular artifact by linear subtraction of wavelet-smoothed vertical and horizontal EOG signals, b) control of muscle artifact by inhibition of BCI training during periods of relatively high power in the 40-64 Hz band. The second BCI, Think Pointer, is a system for two-dimensional cursor control. Steady-state visual evoked potentials (SSVEP) are triggered by four flickering checkerboard stimuli located in narrow strips at each edge of the display. The user attends to one of the four beacons to initiate motion in the desired direction. The SSVEP signals are recorded from eight electrodes located over the occipital region. A KPLS classifier is individually calibrated to map multichannel frequency bands of the SSVEP signals to right-left or up-down motion of a cursor on a computer display. The display stops moving when the user attends to a central fixation point. As for Target Practice, Think Pointer also implements wavelet-based online removal of ocular artifact; however, in Think Pointer muscle artifact is controlled via adaptive normalization of the SSVEP. Training of the classifier requires about three minutes. We have tested our system in real-time operation in three human subjects. Across subjects and sessions, control accuracy ranged from 80% to 100% correct with lags of 1-5 seconds for movement initiation and turning.

Trejo, Leonard J.↗