Search NASA⌕ Search

SEARCH · Search NASA

Results for “dimension reduction”

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 19 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS2000

Dimension Reduction of Hyperspectral Data on Beowulf Clusters

Traditional remote sensing instruments are multispectral, where observations are collected at a few different spectral bands. Recently, many hyperspectral instruments, that can collect observations at hundreds of bands, have been operation. Furthermore, there have been ongoing research efforts on ultraspectral instruments that can produce observations at thousands of spectral bands. While these remote sensing technology developments hold a great promise for new findings in the area of Earth and space science, they present many challenges. These include the need for faster processing of such increased data volumes, and methods for data reduction. Dimension Reduction is a spectral transformation, which is used widely in remote sensing, is the Principal Components Analysis (PCA). In light of the growing number of spectral channels of modern instruments, the paper reports on the development of a parallel PCA and its implementation on two Beowulf cluster configurations, on with fast Ethernet switch and the other is with a Myrinet interconnection.

El-Ghazawi, Tarek↗
NASA NTRS1978

Linear dimension reduction and Bayes classification

An explicit expression for a compression matrix T of smallest possible left dimension K consistent with preserving the n variate normal Bayes assignment of X to a given one of a finite number of populations and the K variate Bayes assignment of TX to that population was developed. The Bayes population assignment of X and TX were shown to be equivalent for a compression matrix T explicitly calculated as a function of the means and covariances of the given populations.

Decell, H. P., Jr.↗
NASA NTRS2014

Neural Network Machine Learning and Dimension Reduction for Data Visualization

Neural network machine learning in computer science is a continuously developing field of study. Although neural network models have been developed which can accurately predict a numeric value or nominal classification, a general purpose method for constructing neural network architecture has yet to be developed. Computer scientists are often forced to rely on a trial-and-error process of developing and improving accurate neural network models. In many cases, models are constructed from a large number of input parameters. Understanding which input parameters have the greatest impact on the prediction of the model is often difficult to surmise, especially when the number of input variables is very high. This challenge is often labeled the "curse of dimensionality" in scientific fields. However, techniques exist for reducing the dimensionality of problems to just two dimensions. Once a problem's dimensions have been mapped to two dimensions, it can be easily plotted and understood by humans. The ability to visualize a multi-dimensional dataset can provide a means of identifying which input variables have the highest effect on determining a nominal or numeric output. Identifying these variables can provide a better means of training neural network models; models can be more easily and quickly trained using only input variables which appear to affect the outcome variable. The purpose of this project is to explore varying means of training neural networks and to utilize dimensional reduction for visualizing and understanding complex datasets.

Liles, Charles A.↗
NASA NTRS2018

An Active Subspace Method for Accelerating Convergence in Delaunay-Based Optimization via Dimension Reduction

Delaunay-based derivative-free optimization, ∆DOGS, is an efficient and provably-convergent global optimization method for the problems which has computationally expensive objection function and the analytical expression for the objective function is not available. ∆-DOGS is a novel optimization scheme in the family of response surface methods (RSMs); however, it suffers from the curse of dimensionality since the computational cost increases dramatically as the number of design parameters increases. As a result, the number of design parameters in ∆-DOGS algorithm is relatively low (n.10). To avoid such problems, this paper proposes a combination of derivative-free optimization, seeking the global minimizer of an expensive and nonconvex objective function f(x) and active subspace method, detecting the directions of the most variability using evaluations of the gradient. The contribution of other directions to the objective function is bounded by a sufficiently small constant. This new algorithm iteratively applied Delaunay-based derivative-free optimization to seek the minimizer on the d-dimensional active subspace that has most function variation. Inverse mapping is needed to project data from active subspace to full-model for evaluating function values. This task is overcome by solving an inequality constrained problem that curves the response surface of the objective function. The test results show that this strategy is effective on a handful of optimization problems.

Bewley, Thomas R.↗
NASA NTRS1998

High Performance Parallel Architectures

Traditional remote sensing instruments are multispectral, where observations are collected at a few different spectral bands. Recently, many hyperspectral instruments, that can collect observations at hundreds of bands, have been operational. Furthermore, there have been ongoing research efforts on ultraspectral instruments that can produce observations at thousands of spectral bands. While these remote sensing technology developments hold great promise for new findings in the area of Earth and space science, they present many challenges. These include the need for faster processing of such increased data volumes, and methods for data reduction. Dimension Reduction is a spectral transformation, aimed at concentrating the vital information and discarding redundant data. One such transformation, which is widely used in remote sensing, is the Principal Components Analysis (PCA). This report summarizes our progress on the development of a parallel PCA and its implementation on two Beowulf cluster configuration; one with fast Ethernet switch and the other with a Myrinet interconnection. Details of the implementation and performance results, for typical sets of multispectral and hyperspectral NASA remote sensing data, are presented and analyzed based on the algorithm requirements and the underlying machine configuration. It will be shown that the PCA application is quite challenging and hard to scale on Ethernet-based clusters. However, the measurements also show that a high- performance interconnection network, such as Myrinet, better matches the high communication demand of PCA and can lead to a more efficient PCA execution.

El-Ghazawi, Tarek↗
NASA NTRS2015

Optimization of Selected Remote Sensing Algorithms for Embedded NVIDIA Kepler GPU Architecture

This paper evaluates the potential of embedded Graphic Processing Units in the Nvidias Tegra K1 for onboard processing. The performance is compared to a general purpose multi-core CPU and full fledge GPU accelerator. This study uses two algorithms: Wavelet Spectral Dimension Reduction of Hyperspectral Imagery and Automated Cloud-Cover Assessment (ACCA) Algorithm. Tegra K1 achieved 51 for ACCA algorithm and 20 for the dimension reduction algorithm, as compared to the performance of the high-end 8-core server Intel Xeon CPU with 13.5 times higher power consumption.

Riha, Lubomir↗
NASA NTRS2015

Optimization of Selected Remote Sensing Algorithms for Embedded NVIDIA Kepler GPU Architecture

This paper evaluates the potential of embedded Graphic Processing Units (GPUs) in Nvidia's Tegra K1 (based on Kepler (TM) hardware) for onboard processing. The performance is compared to a general purpose multi-core CPU (Central Processing Unit) and a fully-fledged GPU accelerator. This study uses two algorithms: Wavelet Spectral Dimension Reduction of Hyperspectral Imagery and Automated Cloud-Cover Assessment (ACCA) Algorithm. Tegra K1 achieved 51 for the ACCA algorithm and 20 for the dimension reduction algorithm, as compared to the performance of the high-end 8-core server Intel Xeon CPU with 13.5 times higher power consumption.

data reduction↗
NASA NTRS2002

Spectral Data Reduction via Wavelet Decomposition

The greatest advantage gained from hyperspectral imagery is that narrow spectral features can be used to give more information about materials than was previously possible with broad-band multispectral imagery. For many applications, the new larger data volumes from such hyperspectral sensors, however, present a challenge for traditional processing techniques. For example, the actual identification of each ground surface pixel by its corresponding reflecting spectral signature is still one of the most difficult challenges in the exploitation of this advanced technology, because of the immense volume of data collected. Therefore, conventional classification methods require a preprocessing step of dimension reduction to conquer the so-called "curse of dimensionality." Spectral data reduction using wavelet decomposition could be useful, as it does not only reduce the data volume, but also preserves the distinctions between spectral signatures. This characteristic is related to the intrinsic property of wavelet transforms that preserves high- and low-frequency features during the signal decomposition, therefore preserving peaks and valleys found in typical spectra. When comparing to the most widespread dimension reduction technique, the Principal Component Analysis (PCA), and looking at the same level of compression rate, we show that Wavelet Reduction yields better classification accuracy, for hyperspectral data processed with a conventional supervised classification such as a maximum likelihood method.

Kaewpijit, S.↗
NASA NTRS2004

Error Analysis for Discontinuous Galerkin Method for Parabolic Problems

In the proposal, the following three objectives are stated: (1) A p-version of the discontinuous Galerkin method for a one dimensional parabolic problem will be established. It should be recalled that the h-version in space was used for the discontinuous Galerkin method. An a priori error estimate as well as a posteriori estimate of this p-finite element discontinuous Galerkin method will be given. (2) The parameter alpha that describes the behavior double vertical line u(sub t)(t) double vertical line 2 was computed exactly. This was made feasible because of the explicitly specified initial condition. For practical heat transfer problems, the initial condition may have to be approximated. Also, if the parabolic problem is proposed on a multi-dimensional region, the parameter alpha, for most cases, would be difficult to compute exactly even in the case that the initial condition is known exactly. The second objective of this proposed research is to establish a method to estimate this parameter. This will be done by computing two discontinuous Galerkin approximate solutions at two different time steps starting from the initial time and use them to derive alpha. (3) The third objective is to consider the heat transfer problem over a two dimensional thin plate. The technique developed by Vogelius and Babuska will be used to establish a discontinuous Galerkin method in which the p-element will be used for through thickness approximation. This h-p finite element approach, that results in a dimensional reduction method, was used for elliptic problems, but the application appears new for the parabolic problem. The dimension reduction method will be discussed together with the time discretization method.

Kaneko, Hideaki↗
NASA NTRS1973

An Iterative Approach to the Feature Selection Problem

The problem dealt with concerns feature selection or reducing the dimension of the data to be processed from n to k. By reducing the dimension of the data from n to k, classification time is generally reduced. Yet the dimension reduction should not be so great that classification accuracy is impaired. Thus, the general problem is considered of classifying an n-dimensional observation vector x into one of m-distinct classes where each class is normally distributed with mean and covariance. It is shown that the probability of misclassification is minimized if a maximum likelihood classification procedure is used to classify the data. The dimension of each observation vector to be processed is conveniently reduced by performing the transformation y = Bx, where B is a K by n matrix of rank k. Thus, the n-dimensional classification problem transforms into a k-dimensional classification problem.

Decell, H. P., Jr.↗
NASA NTRS1988

Planetary research at Lowell Observatory

Scientific goals include a better determination of the basic physical characteristics of cometary nuclei, a more complete understanding of the complex processes in the comae, a survey of abundances and gas/dust ratios in a large number of comets, and measurement of primordial (12)C/(13)C and (14)N/(15)N ratios. The program also includes the observation of Pluto-Charon mutual eclipses to derive dimensions. Reduction and analysis of extensive narrowband photometry of Comet Halley from Cerro Tololo Inter-American Observatory, Perth Observatory, Lowell Observatory, and Mauna Kea Observatory were completed. It was shown that the 7.4-day periodicity in the activity of Comet Halley was present from late February through at least early June 1986, but there is no conclusive evidence of periodic variability in the preperihelion data. Greatly improved NH scalelengths and lifetimes were derived from the Halley data which lead to the conclusion that the abundance of NH in comets is much higher than previously believed. Simultaneous optical and thermal infrared observations were obtained of Comet P/Temple 2 using the MKO 2.2 m telescope and the NASA IRTF. Preliminary analysis of these observations shows that the comet's nucleus is highly elongated, very dark, and quite red.

Baum, William A.↗
NASA NTRS1999

Modeling of polychromatic attenuation using computed tomography reconstructed images

This paper presents a procedure for estimating an accurate model of the CT imaging process including spectral effects. As raw projection data are typically unavailable to the end-user, we adopt a post-processing approach that utilizes the reconstructed images themselves. This approach includes errors from x-ray scatter and the nonidealities of the built-in soft tissue correction into the beam characteristics, which is crucial to beam hardening correction algorithms that are designed to be applied directly to CT reconstructed images. We formulate this approach as a quadratic programming problem and propose two different methods, dimension reduction and regularization, to overcome ill conditioning in the model. For the regularization method we use a statistical procedure, Cross Validation, to select the regularization parameter. We have constructed step-wedge phantoms to estimate the effective beam spectrum of a GE CT-I scanner. Using the derived spectrum, we computed the attenuation ratios for the wedge phantoms and found that the worst case modeling error is less than 3% of the corresponding attenuation ratio. We have also built two test (hybrid) phantoms to evaluate the effective spectrum. Based on these test phantoms, we have shown that the effective beam spectrum provides an accurate model for the CT imaging process. Last, we used a simple beam hardening correction experiment to demonstrate the effectiveness of the estimated beam profile for removing beam hardening artifacts. We hope that this estimation procedure will encourage more independent research on beam hardening corrections and will lead to the development of application-specific beam hardening correction algorithms.

NASA Discipline Musculoskeletal↗
NASA NTRS2004

An Empirical Cumulus Parameterization Scheme for a Global Spectral Model

Realistic vertical heating and drying profiles in a cumulus scheme is important for obtaining accurate weather forecasts. A new empirical cumulus parameterization scheme based on a procedure to improve the vertical distribution of heating and moistening over the tropics is developed. The empirical cumulus parameterization scheme (ECPS) utilizes profiles of Tropical Rainfall Measuring Mission (TRMM) based heating and moistening derived from the European Centre for Medium- Range Weather Forecasts (ECMWF) analysis. A dimension reduction technique through rotated principal component analysis (RPCA) is performed on the vertical profiles of heating (Q1) and drying (Q2) over the convective regions of the tropics, to obtain the dominant modes of variability. Analysis suggests that most of the variance associated with the observed profiles can be explained by retaining the first three modes. The ECPS then applies a statistical approach in which Q1 and Q2 are expressed as a linear combination of the first three dominant principal components which distinctly explain variance in the troposphere as a function of the prevalent large-scale dynamics. The principal component (PC) score which quantifies the contribution of each PC to the corresponding loading profile is estimated through a multiple screening regression method which yields the PC score as a function of the large-scale variables. The profiles of Q1 and Q2 thus obtained are found to match well with the observed profiles. The impact of the ECPS is investigated in a series of short range (1-3 day) prediction experiments using the Florida State University global spectral model (FSUGSM, T126L14). Comparisons between short range ECPS forecasts and those with the modified Kuo scheme show a very marked improvement in the skill in ECPS forecasts. This improvement in the forecast skill with ECPS emphasizes the importance of incorporating realistic vertical distributions of heating and drying in the model cumulus scheme. This also suggests that in the absence of explicit models for convection, the proposed statistical scheme improves the modeling of the vertical distribution of heating and moistening in areas of deep convection.

Rajendran, K.↗
NASA NTRS2004

Registration and Fusion of Multiple Source Remotely Sensed Image Data

Earth and Space Science often involve the comparison, fusion, and integration of multiple types of remotely sensed data at various temporal, radiometric, and spatial resolutions. Results of this integration may be utilized for global change analysis, global coverage of an area at multiple resolutions, map updating or validation of new instruments, as well as integration of data provided by multiple instruments carried on multiple platforms, e.g. in spacecraft constellations or fleets of planetary rovers. Our focus is on developing methods to perform fast, accurate and automatic image registration and fusion. General methods for automatic image registration are being reviewed and evaluated. Various choices for feature extraction, feature matching and similarity measurements are being compared, including wavelet-based algorithms, mutual information and statistically robust techniques. Our work also involves studies related to image fusion and investigates dimension reduction and co-kriging for application-dependent fusion. All methods are being tested using several multi-sensor datasets, acquired at EOS Core Sites, and including multiple sensors such as IKONOS, Landsat-7/ETM+, EO1/ALI and Hyperion, MODIS, and SeaWIFS instruments. Issues related to the coregistration of data from the same platform (i.e., AIRS and MODIS from Aqua) or from several platforms of the A-train (i.e., MLS, HIRDLS, OMI from Aura with AIRS and MODIS from Terra and Aqua) will also be considered.

LeMoigne, Jacqueline↗
NASA NTRS1987

Effects of transverse temperature field nonuniformity on stress in silicon sheet growth

Stress and strain rate distributions are calculated using finite element analysis for steady-state growth of thin silicon sheet temperature nonuniformities imposed in the transverse (sheet width) dimension. Significant reductions in residual stress are predicted to occur for the case where the sheet edge is cooled relative to its center provided plastic deformation with high creep rates is present.

Mataga, P. A.↗
NASA NTRS1991

The attractor dimension of solar decimetric radio pulsations

The temporal characteristics of decimetric pulsations and related radio emissions during solar flares are analyzed using statistical methods recently developed for nonlinear dynamic systems. The results of the analysis is consistent with earlier reports on low-dimensional attractors of such events and yield a quantitative description of their temporal characteristics and hidden order. The estimated dimensions of typical decimetric pulsations are generally in the range of 3.0 + or - 0.5. Quasi-periodic oscillations and sudden reductions may have dimensions as low as 2. Pulsations of decimetric type IV continua have typically a dimension of about 4.

Kurths, J.↗