Image Adaptive High Volume Data Hiding Based on Scalar Quantization

Information-theoretic analysis for data hiding prescribe embedding the hidden data in the choice of quantizer for the host data. In this paper, we consider a suboptimal implementation of this prescription, with a view to hiding high volumes of data in images with low perceptual degradation. The three main findings are as follows: (i) Scalar quantization based data hiding schemes incur about 2 dB penalty from the optimal embedding strategy, which involves vector quantization of the host. (ii) In order to limit perceivable distortion while hiding large amounts of data, hiding schemes must use local perceptual criteria in addition to information-theoretic guidelines. (iii) Powerful erasure and error correcting codes provide a flexible framework that allows the data-hider freedom of choice of where to embed without requiring synchronization between encoder and decoder.