Reversible Data Hiding Method Based on Combining IPVO with Bias-Added Rhombus Predictor by Multi-Predictor Mechanism

Guojun Fan,Zhibin Pan,Erdun Gao,Xinyi Gao,Xiaoran Zhang
DOI: https://doi.org/10.1016/j.sigpro.2020.107888
IF: 4.729
2021-01-01
Signal Processing
Abstract:The well-known pixel-value-ordering (PVO) framework based reversible data hiding (RDH) methods like improved pixel-value-ordering (IPVO) take fully advantages of sorting and achieve an outstanding performance. In this paper, we merge the advantages of sorting and averaging by combining IPVO and conventional Rhombus predictor through the multi-predictor mechanism. Furthermore, in order to better utilize the advantages of multi-predictor mechanism, a bias is added to improve the conventional Rhombus predictor. By adding adaptive biases for the pixels with different complexities, our proposed bias-added Rhombus predictor can reduce the embedding distortion significantly. Moreover, a two-step optimization is proposed to save the running time. Experimental results verify that the proposed method outperforms state-of-the-art PVO based RDH methods. (C) 2020 Elsevier B.V. All rights reserved.
What problem does this paper attempt to address?