A novel reversible data hiding scheme based on difference-histogram modification and optimal EMD algorithm

• Our proposed scheme can embed a large amount of secrets without dramatically distortion.
• Construct an optimal exploiting modification direction matrix for embedding data.
• Experiments show our scheme has a better performance compared with other schemes.

Data hiding techniques, which are classified into two groups: irreversible and reversible techniques, have been gained more and more attention recently. In which, reversible data hiding (RDH) is paid more attention since it is capable to recover original host images after revealing the secret data. However, there always is tradeoff between embedding capacity and image quality. The more secret data is embedded, the lower image quality is obtained. Therefore, in this paper, we propose a new RDH scheme to embed a large amount of secrets without causing dramatically damage to images. We develop an optimal exploiting modification direction matrix to modify the difference histogram. By carefully selecting image regions and using the optimal EMD table, the output image quality can be preserved while the embedding capacity of the scheme can be significantly enhanced. Experimental results demonstrated that our scheme outperforms previous schemes in terms of embedding capacity and image quality.