Reversible data hiding for encrypted images with improved performance is introduced in this paper. Each unit in the original image is separated into three components, and each component is encrypted by Paillier homomorphic encryption. Additional bits can be concealed into the encrypted image by manipulating the encrypted signals. Finally, the original image is obtained without error when the direct decryption is applied. The embedded bits are perfectly extracted as well. Optimal visual quality and improved embedding rate are obtained by the proposed approach, since the value of the directly decrypted unit is the same as the original one. Experimental results and comparisons are demonstrated to illustrate the effectiveness and advantages of the proposed method.
CITATION STYLE
Wu, X., Liang, Z., Chen, B., & Liu, T. (2016). On improving homomorphic encryption-based reversible data hiding. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10039 LNCS, pp. 28–38). Springer Verlag. https://doi.org/10.1007/978-3-319-48671-0_3
Mendeley helps you to discover research relevant for your work.