Recognizing shipbuilding parts using artificial neural networks and Fourier descriptors

    Research output: Contribution to journalArticlepeer-review

    118 Downloads (Pure)

    Abstract

    A pattern recognition system is described for recognizing shipbuilding parts using artificial neural networks and Fourier descriptors. The system uses shape contour information that is invariant of size, translation, and rotation. Fourier descriptors provide information, and the neural networks make decisions about the shapes. A brief review of the current state of the art is included, and results from testing show that the system distinguished between various shapes and proved to be a valid approach.
    Original languageEnglish
    Pages (from-to)337-342
    Number of pages6
    JournalProceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture
    Volume223
    Issue number3
    DOIs
    Publication statusPublished - 1 Mar 2009

    Fingerprint

    Dive into the research topics of 'Recognizing shipbuilding parts using artificial neural networks and Fourier descriptors'. Together they form a unique fingerprint.

    Cite this