An Information-Theoretic Model for Image Watermarking and Data Hiding

An information?theoretic model for image watermarking and data hiding systems is presented in this paper. The fundamental capacity limits of these systems are determined by the statistical model used for the host image, by the distortion constraints on the data hider and the attacker, and by the information available to the data hider, to the attacker, and to the decoder. We consider wavelet statistical models for images and compute data-hiding capacity for compressed and uncompressed host-image sources. Closedform expressions are obtained under sparse-model approximations.

• Type: Conference
• Year: 2000
• Where: ICIP
• Authors: Gen-Iu. Lin, Mehmet Kivanç Mihçak, Pierre Moulin
• Keywords: Data Hider, Data Hiding Systems, Fundamental Capacity Limits, ICIP 2000, Image Processing