Separable robust data hiding in encrypted image based on continuous quadrant tree and 2Bin N-nary

Hui Shi, Baoyue Hu, Meihan Chen, Yanni Li, Yonggong Ren

Abstract: Data hiding is a noteworthy research topic in digital technology for years. It can be used for copyright protection, authentication, content-ownership verification, and sending patient information. Focusing on how to improve the security, capacity, visual quality and robustness, a separable robust data hiding in encrypted image based on Continuous Quadrant Tree (CQT) and 2Bin N-nary is proposed. In this work, the definition of CQT, constraints of the sub-tree roots, generation and traversal of CQT are proposed for the first time. To achieve security, matrix traversal, CQT and rotation encryption are combined. When select embedding interval, a histogram preprocessing algorithm based on pixel fluctuation is presented. This paper achieves separable decryption and extraction. Experimental results demonstrate that the capacity has been improved by 50–70% compared with the similar scheme. It not only achieves high security and high visual quality, but also is robust to various attacks.

External IDs:dblp:journals/paa/ShiHCLR23