High capacity reversible data hiding based on histogram shifting and non-local means

In this paper we propose a new reversible data hiding framework which applies prediction in the embedding procedure by suitably modifying the prediction errors and exploits non-local similarity in the prediction phase to estimate the to-be-predicted value. This results in a scheme which can be jointly used with different predictors and allows reaching high embedding capacity while preserving a high image quality. Extensive simulations demonstrate the effectiveness of the proposed approach.