Color normalization for appearance based recognition of video key-frames

Juan M. Sánchez, Xavier Binefa

    Research output: Contribution to journalArticleResearchpeer-review

    4 Citations (Scopus)

    Abstract

    Color image variations produced by the change of video acquisition devices strongly affect color appearance based recognition tasks in digital video applications. Up to date, many algorithms have been developed in order to achieve independence from different image formation factors like intensity, orientation and color of the illumination source. In this paper, we check the capability of three of these approaches to generate device independent color representations in order to be used as a normalization step previous to the recognition process. Our recently developed color correction approach for video imagery based on Gaussian mixture models has also been tested. Experimental results show that a simple algorithm like the grayworld correction is enough for appearance based recognition under this particular kind of color variability. Similar recognition rates are obtained using our correction, with the advantage that the visual quality of normalized images is preserved. In digital video library applications, where not only recognition tasks are performed, this approach becomes much more useful. © 2000 IEEE.
    Original languageEnglish
    Pages (from-to)815-818
    JournalProceedings - International Conference on Pattern Recognition
    Volume15
    Issue number1
    Publication statusPublished - 1 Dec 2000

    Fingerprint Dive into the research topics of 'Color normalization for appearance based recognition of video key-frames'. Together they form a unique fingerprint.

    Cite this