Reversible Data Hiding Exploiting Variance in Wavelet Coefficients

In this paper, we present a new reversible data hiding scheme that utilizes the wavelet transform and better exploits the large wavelet coefficient variance to achieve high capacity and imperceptible embedding. Our scheme differs from those of previous studies in that the wavelet coefficients histogram rather than the gray-level histogram is manipulated. In addition, we design intelligent histogram-shifting rules to avoid the decimal problem in grayscale pixel values after recovery process to achieve reversibility. Small changes in the wavelet coefficients after embedding process are important factors contributing to low visual distortion in the marked image. Furthermore, an important property of our scheme is that the use of threshold differs greatly from previous schemes. The experimental results show that our scheme outperforms other reversible data hiding schemes.