Information-theoretic analysis of steganalysis in real images
MM&Sec '06 Proceedings of the 8th workshop on Multimedia and security
Introducing a Two Dimensional Measure for Watermarking Capacity in Images
ICIAR '08 Proceedings of the 5th international conference on Image Analysis and Recognition
| Hi-index | 35.68 |
In this paper, the problem of capacity analysis of data-hiding techniques in a game information-theoretic framework is considered. Capacity is determined by the stochastic model of the host image, by the distortion constraints, and by the side information about the watermarking channel state available at the encoder and at the decoder. The importance of the proper modeling of image statistics is emphasized, and for this purpose, a novel stochastic nonstationary image model is proposed that is based on geometrical priors, the so-called edge process model. Being mathematically simple and tractable, the edge process model outperforms the estimation-quantization (EQ) and spike process models in reference applications such as denoising. Finally, this model allows us to obtain a realistic estimate of maximal embedding rates, and in particular, it is shown that the expected capacity limit of real images is significantly lower than previously reported.