Unified entropy-based sorting for reversible data hiding

Reversible data hiding schemes compete against each other for a sharply distributed prediction error histogram, usually realized by utilizing prediction strategies together with sorting technique that aims to estimate the local context complexity for each pixel to optimize the embedding order. Sorting techniques benefit prediction a lot by picking out pixels located in smooth areas. In this paper, a novel entropy-based sorting (EBS) scheme is proposed for reversible data hiding, which uses entropy measurement to characterize local context complexity for each image pixel. Futhermore, by extending the EBS technique to the two-dimensional case, it shows generalized abilities for multi-dimensinal RDH scenarios. Additionally, a new gradient-based tracking and weighting (GBTW) pixel prediction method is introduced to be combined with the EBS technique. Experimental results apparently indicate that our proposed method outperforms the previous state-of-arts counterparts significantly in terms of both the prediction accuracy and the overall embedding performance.