PROJECT TITLE:

Minimum Rate Prediction and Optimized Histograms Modification for Reversible Data Hiding

ABSTRACT:

Prediction-error enlargement (PEE)-based reversible knowledge hiding schemes consist of two steps. 1st, a sharp prediction-error (PE) histogram is generated by utilizing pixel prediction methods. Second, secret messages are reversibly embedded into the prediction-errors through expanding and shifting the PE histogram. Previous PEE methods treat the two steps independently whereas they either specialise in pixel prediction to obtain a sharp PE histogram, or aim at histogram modification to reinforce the embedding performance for a given PE histogram. This paper propose a pixel prediction methodology primarily based on the minimum rate criterion for reversible knowledge hiding, that establishes the consistency between the two steps in essence. And correspondingly, a novel optimized histograms modification theme is presented to approximate the optimal embedding performance on the generated PE sequence. Experiments demonstrate that the proposed technique outperforms the previous state-of-art counterparts considerably in terms of each the prediction accuracy and the final embedding performance.