High-Fidelity Reversible Data Hiding Algorithm Based on SVD

Abstract

The conventional reversible data hiding (RDH) is limited by the problem of distortion caused by embedding the secret data on the original carrier image. Therefore, in this work, a high-fidelity RDH algorithm was proposed based on Singular Value Decomposition (SVD) prediction. Specifically, the target pixel’s neighborhoods were processed under SVD to predict the target pixel. The prediction error histogram is obtained by subtracting the predicted pixel value from the original pixel value. And then, the histogram shifting (HS) strategies is utilized to embed secret data. Extensive experiments have shown that when the embedded capacity is 10000 bits, the Lena’s PSNR was 63.00 dB, which outperforms the current state-of-the-art RDH method with high-fidelity.