An Efficient Perceptual Watermarking Scheme using a New Just-Noticeable-Difference Model Phi Bang Nguyen ∗ , Azeddine Beghdadi, Marie Luong L2TI Laboratory, Galilee Institute, University Paris 13, 99 Ave. J. B. Clment, 93430 Villetaneuse, France Abstract In this paper, a new watermarking scheme using human visual system (HVS) modeling is proposed. The approach consists in building computational mod- els which take into account the most common properties of the HVS that can be exploited for watermarking. Two schemes for embedding and controlling the transparency of the watermark are presented: the implicit and the ex- plicit embedding scheme. Both schemes are designed in the framework of the pyramid decomposition which has been shown to be a powerful tool for ana- lyzing image through a multi-scale representation. For the first approach, a Multi-scale Visibility Map is used to optimize the watermark embedding pro- cess. The second approach makes use of HVS’s properties in an explicit and more sophisticated manner that consists in tuning the watermark strength just beneath the visual detection threshold. A new JND (Just-Noticeable- Difference) model for determining this threshold is then proposed and evalu- ated. The obtained results provide strong support for this new JND model. Keywords: Human Visual System; Watermarking; Perceptual Model; Just Noticeable Difference; Visual Masking; CSF. 1. Introduction For many years, HVS modeling has become an important issue in im- age and multimedia processing such as image compression, quality assess- ment, watermarking and so on. In compression, the limitation of HVS can * The first author is currently affiliated with Technicolor R&D (formerly Thomson), 975 Les Champs Blancs, 35510 Cesson-Sevigne, France Preprint submitted to Signal Processing: Image Communication November 18, 2012 Manuscript NGuyen-Beghdadi- Luong