Optimal Class Separation in Hyperspectral Image Data: Iterated Canonical Discriminant Analysis

Allan Aasbjerg Nielsen, Andreas Müller

    Research output: Contribution to conferencePaperResearchpeer-review

    Abstract

    This paper describes canonical discriminant analysis and sketches an iterative version which is then applied to obtain optimal separation between a region, here examplified by either “water” or “wood/trees” and the rest of a HyMap image. We show that the iterative version greatly enhances the separation between the regions.
    Original languageEnglish
    Publication date2012
    Number of pages4
    Publication statusPublished - 2012
    EventThird Annual Hyperspectral Imaging Conference (HSI2012) - , Italy
    Duration: 15 May 201216 May 2012
    Conference number: 3
    http://istituto.ingv.it/l-ingv/convegni-e-seminari/archivio-congressi/convegni-2012/hsi2012-1/hsi2012/view

    Conference

    ConferenceThird Annual Hyperspectral Imaging Conference (HSI2012)
    Number3
    Country/TerritoryItaly
    Period15/05/201216/05/2012
    Internet address

    Keywords

    • Two-class discrimination
    • HyMap

    Fingerprint

    Dive into the research topics of 'Optimal Class Separation in Hyperspectral Image Data: Iterated Canonical Discriminant Analysis'. Together they form a unique fingerprint.

    Cite this