Loading...

Adaptive watermarking scheme based on ICA and RDWT

Ghaedi Oskooei, S ; Sharif University of Technology

721 Viewed
  1. Type of Document: Article
  2. DOI: 10.1049/ic.2009.0239
  3. Abstract:
  4. This paper proposes a new approach to watermarking multimedia products based on the combination of redundant discrete wavelet transform and independent component analysis. The original image is decomposed by RDWT, and watermark is embedded in to LL sub-band frequency according mixing model of the ICA, after that for enhancing the robustness of the watermark, the perceptual model is applied via stochastic approach for watermark adapting. This is based on computation of a noise visibility function (NVF) which has local image properties so the strength of watermarking is controllable. Principal component analysis (PCA) whitening process and FastICA techniques are introduced to ensure a blind watermark extraction without requiring the original image. Experimental results show the proposed technique is robust under compression attacks
  5. Keywords:
  6. Adaptive watermarking ; Blind ; Blind watermark ; FastICA ; ICA ; Image properties ; Mixing models ; Multimedia products ; New approaches ; Noise visibility function ; NVF ; Original images ; Perceptual model ; RDWT ; Redundant discrete wavelet transform ; Stochastic approach ; Sub-bands ; Whitening process ; Color image processing ; Digital watermarking ; Discrete wavelet transforms ; Principal component analysis ; Stochastic models ; Watermarking ; Independent component analysis
  7. Source: IET Seminar Digest, 3 December 2009 through 3 December 2009 ; Volume 2009, Issue 2 , 2009 ; 9781849192071 (ISBN)
  8. URL: http://ieeexplore.ieee.org/document/5522285