High-Quality Reversible Data Hiding Approach Based on Evaluating Multiple Prediction Methods

Abstract : Reversible data hiding based on prediction methods is a good technique that can hide secret bits into cover images efficiently. In this paper, we propose a reversible data hiding method based on four candidates of prediction methods and local complexity for enhancing stego-image quality. In our proposed method, before we embed the secret message in one level, we evaluate the four prediction methods by calculating their efficiency ratios to decide which prediction method will be used. When the selected prediction method is applied, a threshold based on local complexity is used to determine which pixel should join the shifting and embedding process. Therefore, more pixels will avoid executing the process of pixel shifting. It results in stego-images with lower distortion. The experimental results show that our image quality is superior to that of other approaches at the same capacity.