Improved Reversible Data Hiding Scheme Based on Mean Preservation and Difference Expansion

Abstract

ABSTRACT In this paper, the erroneous scheme proposed by S. Agrawal and M. Kumar, [“Mean value based reversible data hiding in encrypted images,” Optik, Vol. 130, pp. 922–934, 2017.] has been identified, as well as new and improved reversible data hiding (RDH) scheme for digital images has been subsequently proposed. In S. Agrawal and M. Kumar, [“Mean value based reversible data hiding in encrypted images,” Optik, Vol. 130, pp. 922–934, 2017.], they have proposed an RDH scheme based on a floating mean value of a particular row preserved in the embedded image itself. As per the data type standards followed in such technologies, it is impossible to preserve a floating mean value in the embedded image. In this article, by using certain mod operations and offset addition to ensure an integer average value for a specific row, a realizable RDH scheme is proposed. Rows may further be divided into sub-rows of a specific size on the basis of payload required and in such a way that it is a perfect divisor of row size. Corresponding integer mean value is preserved by replacing the first pixel location of the corresponding sub-row. Subsequently, difference expansion scheme is applied further to store the mod value as well as the watermark. Hence, the proposed scheme improves on embedding capacity significantly. Experimental results and performance comparison prove that the proposed method is better than certain existing methods.