• DocumentCode
    1905265
  • Title

    SIFT based geometric distortion correction method

  • Author

    Yavuz, Erkan ; Telatar, Ziya

  • Author_Institution
    Image Process. Dept., Aselsan Electron. Inc., Ankara
  • fYear
    2008
  • fDate
    27-29 Oct. 2008
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    It is very important to cope with geometric distortions especially for robust watermarking systems since these attacks break the synchronization between the watermark and detector. In general, robustness to geometric attacks is considered as a different concept as rotation, scale and translation (RST) invariant watermarking. However, it is inseparable part of a copyright protection system. There are different approaches to overcome this problem. Most of them are good in individual rotation, scaling and translation attacks but it is not the case for combination of them. In this paper, we propose and discuss the performance of a scale invariant feature transform (SIFT) based geometric distortion correction method against rotation, scaling, translation, cropping, flipping and so on. As it restores the approximate original, it can be used in watermarking as a synchronization step.
  • Keywords
    copy protection; geometry; watermarking; SIFT based geometric distortion correction method; copyright protection; invariant watermarking; robust watermarking systems; rotation scale and translation; scale invariant feature transform; Authentication; Copyright protection; Detectors; Forensics; Image coding; Image processing; Image restoration; Internet; Robustness; Watermarking;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer and Information Sciences, 2008. ISCIS '08. 23rd International Symposium on
  • Conference_Location
    Istanbul
  • Print_ISBN
    978-1-4244-2880-9
  • Electronic_ISBN
    978-1-4244-2881-6
  • Type

    conf

  • DOI
    10.1109/ISCIS.2008.4717864
  • Filename
    4717864