ALBERT

All Library Books, journals and Electronic Records Telegrafenberg

X
feed icon rss

Your email was sent successfully. Check your inbox.

An error occurred while sending the email. Please try again.

Proceed reservation?

Export
  • 1
    Electronic Resource
    Electronic Resource
    Clustering with neural networks (1990)
    Springer
    Biological cybernetics 63 (1990), S. 201-208 
    Details
    ISSN: 1432-0770
    Source: Springer Online Journal Archives 1860-2000
    Topics: Biology , Computer Science , Physics
    Notes: Abstract Partitioning a set ofN patterns in ad-dimensional metric space intoK clusters — in a way that those in a given cluster are more similar to each other than the rest — is a problem of interest in many fields, such as, image analysis, taxonomy, astrophysics, etc. As there are approximatelyK N/K! possible ways of partitioning the patterns amongK clusters, finding the best solution is beyond exhaustive search whenN is large. We show that this problem, in spite of its exponential complexity, can be formulated as an optimization problem for which very good, but not necessarily optimal, solutions can be found by using a Hopfield model of neural networks. To obtain a very good solution, the network must start from many randomly selected initial states. The network is simulated on the MPP, a 128 × 128 SIMD array machine, where we use the massive parallelism not only in solving the differential equations that govern the evolution of the network, but also in starting the network from many initial states at once thus obtaining many solutions in one run. We achieve speedups of two to three orders of magnitude over serial implementations and the promise through Analog VLSI implementations of further speedups of three to six orders of magnitude.
    Type of Medium: Electronic Resource
    URL: http://dx.doi.org/10.1007/BF00195859
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    Paper (German National Licenses)
    Fulltext
  • 2
    facet.materialart.
    Unknown
    The Feynman Propagator from a Single Path (2002)
    American Physical Society
    Details
    Publication Date: 2002-12-02
    Print ISSN: 0031-9007
    Electronic ISSN: 1079-7114
    Topics: Physics
    Published by American Physical Society
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    PAPER CURRENT
  • 3
    facet.materialart.
    Unknown
    Clustering with neural networks (1990)
    Springer
    Details
    Publication Date: 1990-07-01
    Print ISSN: 0340-1200
    Electronic ISSN: 1432-0770
    Topics: Biology , Computer Science , Physics
    Published by Springer
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    PAPER CURRENT
  • 4
    facet.materialart.
    Unknown
    Phase equilibria in polydisperse fluids (1982)
    American Institute of Physics
    Details
    Publication Date: 1982-07-01
    Print ISSN: 0021-9606
    Electronic ISSN: 1089-7690
    Topics: Chemistry and Pharmacology , Physics
    Published by American Institute of Physics
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    PAPER CURRENT
  • 5
    facet.materialart.
    Unknown
    Hyperspectral Image Classification using a Self-Organizing Map (2001)
    In:  CASI
    Details
    Publication Date: 2004-12-03
    Description: The use of hyperspectral data to determine the abundance of constituents in a certain portion of the Earth's surface relies on the capability of imaging spectrometers to provide a large amount of information at each pixel of a certain scene. Today, hyperspectral imaging sensors are capable of generating unprecedented volumes of radiometric data. The Airborne Visible/Infrared Imaging Spectrometer (AVIRIS), for example, routinely produces image cubes with 224 spectral bands. This undoubtedly opens a wide range of new possibilities, but the analysis of such a massive amount of information is not an easy task. In fact, most of the existing algorithms devoted to analyzing multispectral images are not applicable in the hyperspectral domain, because of the size and high dimensionality of the images. The application of neural networks to perform unsupervised classification of hyperspectral data has been tested by several authors and also by us in some previous work. We have also focused on analyzing the intrinsic capability of neural networks to parallelize the whole hyperspectral unmixing process. The results shown in this work indicate that neural network models are able to find clusters of closely related hyperspectral signatures, and thus can be used as a powerful tool to achieve the desired classification. The present work discusses the possibility of using a Self Organizing neural network to perform unsupervised classification of hyperspectral images. In sections 3 and 4, the topology of the proposed neural network and the training algorithm are respectively described. Section 5 provides the results we have obtained after applying the proposed methodology to real hyperspectral data, described in section 2. Different parameters in the learning stage have been modified in order to obtain a detailed description of their influence on the final results. Finally, in section 6 we provide the conclusions at which we have arrived.
    Keywords: Earth Resources and Remote Sensing
    Type: Proceedings of the Tenth JPL Airborne Earth Science Workshop; 267-274
    Format: text
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    NASA TECHNICAL REPORTS
  • 6
    facet.materialart.
    Unknown
    Neural networks for atmospheric retrievals (1993)
    In:  CASI
    Details
    Publication Date: 2013-08-31
    Description: We use neural networks to perform retrievals of temperature and water fractions from simulated clear air radiances for the Atmospheric Infrared Sounder (AIRS). Neural networks allow us to make effective use of the large AIRS channel set, and give good performance with noisy input. We retrieve surface temperature, air temperature at 64 distinct pressure levels, and water fractions at 50 distinct pressure levels. Using 728 temperature and surface sensitive channels, the RMS error for temperature retrievals with 0.2K input noise is 1.2K. Using 586 water and temperature sensitive channels, the mean error with 0.2K input noise is 16 percent. Our implementation of backpropagation training for neural networks on the 16,000-processor MasPar MP-1 runs at a rate of 90 million weight updates per second, and allows us to train large networks in a reasonable amount of time. Once trained, the network can be used to perform retrievals quickly on a workstation of moderate power.
    Keywords: CYBERNETICS
    Type: The 1993 Goddard Conference on Space Applications of Artificial Intelligence; p 155-167
    Format: application/pdf
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    NASA TECHNICAL REPORTS
  • 7
    facet.materialart.
    Unknown
    How to cluster in parallel with neural networks (1988)
    In:  CASI
    Details
    Publication Date: 2013-08-31
    Description: Partitioning a set of N patterns in a d-dimensional metric space into K clusters - in a way that those in a given cluster are more similar to each other than the rest - is a problem of interest in astrophysics, image analysis and other fields. As there are approximately K(N)/K (factorial) possible ways of partitioning the patterns among K clusters, finding the best solution is beyond exhaustive search when N is large. Researchers show that this problem can be formulated as an optimization problem for which very good, but not necessarily optimal solutions can be found by using a neural network. To do this the network must start from many randomly selected initial states. The network is simulated on the MPP (a 128 x 128 SIMD array machine), where researchers use the massive parallelism not only in solving the differential equations that govern the evolution of the network, but also by starting the network from many initial states at once, thus obtaining many solutions in one run. Researchers obtain speedups of two to three orders of magnitude over serial implementations and the promise through Analog VLSI implementations of speedups comensurate with human perceptual abilities.
    Keywords: COMPUTER SYSTEMS
    Type: The 2nd Symposium on the Frontiers of Massively Parallel Computations; p 31-38
    Format: application/pdf
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    NASA TECHNICAL REPORTS
  • 8
    facet.materialart.
    Unknown
    Detection of edges using local geometry (1989)
    In:  CASI
    Details
    Publication Date: 2013-08-31
    Description: Researchers described a new representation, the local geometry, for early visual processing which is motivated by results from biological vision. This representation is richer than is often used in image processing. It extracts more of the local structure available at each pixel in the image by using receptive fields that can be continuously rotated and that go to third order spatial variation. Early visual processing algorithms such as edge detectors and ridge detectors can be written in terms of various local geometries and are computationally tractable. For example, Canny's edge detector has been implemented in terms of a local geometry of order two, and a ridge detector in terms of a local geometry of order three. The edge detector in local geometry was applied to synthetic and real images and it was shown using simple interpolation schemes that sufficient information is available to locate edges with sub-pixel accuracy (to a resolution increase of at least a factor of five). This is reasonable even for noisy images because the local geometry fits a smooth surface - the Taylor series - to the discrete image data. Only local processing was used in the implementation so it can readily be implemented on parallel mesh machines such as the MPP. Researchers expect that other early visual algorithms, such as region growing, inflection point detection, and segmentation can also be implemented in terms of the local geometry and will provide sufficiently rich and robust representations for subsequent visual processing.
    Keywords: INSTRUMENTATION AND PHOTOGRAPHY
    Type: NASA, Langley Research Center, Visual Information Processing for Television and Telerobotics; p 109-119
    Format: application/pdf
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    NASA TECHNICAL REPORTS
  • 9
    facet.materialart.
    Unknown
    Distance between images (1992)
    In:  Other Sources
    Details
    Publication Date: 2019-07-13
    Description: Comparing two binary images and assigning a quantitative measure to this comparison finds its purpose in such tasks as image recognition, image compression, and image browsing. This quantitative measurement may be computed by utilizing the Hausdorff distance of the images represented as two-dimensional point sets. In this paper, we review two algorithms that have been proposed to compute this distance, and we present a parallel implementation of one of them on the MasPar parallel processor. We study their complexity and the results obtained by these algorithms for two different types of images: a set of displaced pairs of images of Gaussian densities, and a comparison of a Canny edge image with several edge images from a hierarchical region growing code.
    Keywords: CYBERNETICS
    Type: ; 8 p.|Frontiers 92 - Symposium on the Frontiers of Massively Parallel Computation; Oct 19, 1992 - Oct 21, 1992; McLean, VA; United States
    Format: text
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    NASA TECHNICAL REPORTS
  • 10
    facet.materialart.
    Unknown
    Solving inversion problems with neural networks (1990)
    In:  Other Sources
    Details
    Publication Date: 2019-07-13
    Description: A class of inverse problems in remote sensing can be characterized by Q = F(x), where F is a nonlinear and noninvertible (or hard to invert) operator, and the objective is to infer the unknowns, x, from the observed quantities, Q. Since the number of observations is usually greater than the number of unknowns, these problems are formulated as optimization problems, which can be solved by a variety of techniques. The feasibility of neural networks for solving such problems is presently investigated. As an example, the problem of finding the atmospheric ozone profile from measured ultraviolet radiances is studied.
    Keywords: CYBERNETICS
    Type: IJCNN - International Joint Conference on Neural Networks; Jun 17, 1990 - Jun 21, 1990; San Diego, CA; United States
    Format: text
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
    NASA TECHNICAL REPORTS
Close ⊗
This website uses cookies and the analysis tool Matomo. More information can be found here...