• DocumentCode
    2161698
  • Title

    Color image-adaptive watermarking

  • Author

    Gilani, S.A.M. ; Kostopoulos, I. ; Skodras, A.N.

  • Author_Institution
    Electron. Lab., Patras Univ., Greece
  • Volume
    2
  • fYear
    2002
  • fDate
    2002
  • Firstpage
    721
  • Abstract
    A novel scheme for the watermarking of colour images is presented in this communication. The first objective is to find the most suitable alternative to RGB color space, which is highly correlated. Colour spaces with linear relation to RGB colour space with uncorrelated components are found to be most suitable for watermarking applications. Second objective is to make the scheme adaptive to oˆc$80a˚ colour image. This is achieved by keeping the PSNR in a predefined quality range, while timing the watermark strength parameter. Watermark detection is fast and blind, i.e. only the watermark generation and coefficient randomization keys are needed, and not the original image. The presented results demonstrate the robustness of the method against some common image processing attacks such as compression, scaling, uniform or gaussian noise addition, median filtering, cropping and multiple watermarking.
  • Keywords
    adaptive signal processing; image coding; image colour analysis; watermarking; Gaussian noise addition; PSNR; coefficient randomization keys; colour images; colour spaces; compression; cropping; image processing attacks; median filtering; multiple watermarking; quality; scaling; strength parameter; uncorrelated components; uniform noise addition; watermark detection; watermark generation keys; Color; Discrete cosine transforms; Discrete wavelet transforms; Filtering; Frequency; Laboratories; Low-frequency noise; Robustness; Space technology; Watermarking;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Digital Signal Processing, 2002. DSP 2002. 2002 14th International Conference on
  • Print_ISBN
    0-7803-7503-3
  • Type

    conf

  • DOI
    10.1109/ICDSP.2002.1028192
  • Filename
    1028192