Quadruple histogram shifting-based reversible information hiding approach for digital images

Abstract

Nowadays, data hiding techniques are one of the most growing researching areas. This paper introduces a quadruple histogram shifting-based reversible data embedding approach for digital images. Pixel differences distribution is utilized for maximizing the hiding payload capacity while preserving minimum degradations for stego-images. The improved histogram-based reversible data embedding algorithm is designed with the potential of maximizing the hiding payload capacity while keeping a perfect stego-image quality with data lossless. The proposed algorithm utilizes a 3 × 3 -box predictive filter and can hide 2-bits per pixel to increase the payload capacity. Firstly, an embedding phase has been performed by applying a numerical ordering strategy of the pixels to be predicted by the 3-by-3 box filter. The proposed algorithm is performing this ordering strategy in four stages to avoid affecting the follow-up prediction of all other pixels. In the other side, an extracting phase has been performed by reversing the four stages to generate the embedded data. The comparison with state-of-the-art schemes is performed to validate and guarantee the prevalence of the improved histogram-based reversible data embedding approach. The obtained test results demonstrated and confirmed that the proposed improved histogram-based reversible data embedding approach not only can improve the payload capacity but also can save the stego-image quality.