Analysis of neural networks for edge detection

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    79 Downloads (Pure)

    Abstract

    This paper illustrates a novel method to analyze artificial neural networks so as to gain insight into their internal functionality. To this purpose, the elements of a feedforward-backpropagation neural network, that has been trained to detect edges in images, are described in terms of differential operators of various orders and with various angles of operation.
    Original languageUndefined
    Title of host publication13th Annual Workshop on Circuits Systems and Signal Processing (ProRISC)
    Place of PublicationNetherlands
    PublisherSTW
    Pages580-586
    Number of pages7
    ISBN (Print)90-73461-33-2
    Publication statusPublished - Nov 2002
    Event13th Workshop on Circuits, Systems and Signal Processing, ProRISC 2002 - Veldhoven, Netherlands
    Duration: 28 Nov 200229 Nov 2002
    Conference number: 13

    Publication series

    Name
    PublisherSTW Technology Foundation

    Workshop

    Workshop13th Workshop on Circuits, Systems and Signal Processing, ProRISC 2002
    CountryNetherlands
    CityVeldhoven
    Period28/11/0229/11/02

    Keywords

    • METIS-206911
    • IR-43578
    • Edge detection
    • Neural Networks
    • EWI-1772
    • rule extraction
    • gradient filters
    • digital image processing

    Cite this