Comparison of multispectral images across the Internet

G.W.A.M. van der Heijden, G. Polder, Th. Gevers

    Research output: Chapter in Book/Report/Conference proceedingConference paper

    9 Citations (Scopus)

    Abstract

    Comparison in the RGB domain is not suitable for precise color matching, due to the strong dependency of this domain on factors like spectral power distribution of the light source and object geometry. We have studied the use of multispectral or hyperspectral images for color matching, since it can be proven that hyperspectral images can be made independent of the light source (color constancy) and object geometry (normalized color constancy). Hyperspectral images have the disadvantage that they are large compared to regular RGB-images, which makes it infeasible to use them for image matching across the Internet. For red roses, it is possible to reduce the large number of bands (>100) of the spectral images to only three bands, the same number as of an RGB-image, using Principal Component Analysis, while maintaining 99% of the original variation. The obtained PCA-images of the roses can be matched using for example histogram cross correlation. From the principal coordinates plot, obtained from the histogram similarity matrices of twenty images of red roses, the discriminating power seems to be better for normalized spectral images than for color constant spectral images and RGB-images, the latter being recorded under highly optimized standard conditions.
    Original languageEnglish
    Title of host publicationInternet Imaging, SPIE Conference 26-28 January 2000
    EditorsR. Schettini, G.B. Beretta
    Pages196-206
    Number of pages11
    Volume3964
    Publication statusPublished - 2000
    EventSPIE conference 2000 - San Jose, United States
    Duration: 26 Jan 200028 Jan 2000

    Conference

    ConferenceSPIE conference 2000
    CountryUnited States
    CitySan Jose
    Period26/01/0028/01/00

    Fingerprint Dive into the research topics of 'Comparison of multispectral images across the Internet'. Together they form a unique fingerprint.

  • Cite this

    van der Heijden, G. W. A. M., Polder, G., & Gevers, T. (2000). Comparison of multispectral images across the Internet. In R. Schettini, & G. B. Beretta (Eds.), Internet Imaging, SPIE Conference 26-28 January 2000 (Vol. 3964, pp. 196-206) https://edepot.wur.nl/30203