改进的成对预测误差扩展可逆数据隐藏算法

Improved Reversible Data Hiding Algorithm for Pairwise Prediction Error Expansion

基于成对预测误差扩展是目前一种大容量低失真的可逆数据隐藏技术。误差扩展算法通过生成二维预测误差直方图确定最优扩展路径以完成自适应地修改。中值边缘检测器和菱形预测的方法只考虑了部分被测像素周围的相邻像素,本文考虑了被测像素周围的八个像素值,更好地利用了图像像素之间的相关性,并在此基础上,更精确地区分像素局部区域纹理度的高低,来实现选择像素值的精确度。所提方法在嵌入相同容量的秘密信息时,图像的失真度都优于现有算法,具有更好的视觉质量。

Expansion based on pairwise prediction errors is currently a reversible data hiding technology with large capacity and low distortion. The error expansion algorithm determines the optimal expansion path by generating a two-dimensional prediction error histogram to complete the adaptive modifi-cation. The median edge detector and diamond prediction method only considers the neighboring pixels around part of the measured pixel. This article considers the eight pixel values around the measured pixel, which makes better use of the correlation between image pixels, and on this basis, the level of texture in the local area of the pixel is more accurately distinguished to achieve the accuracy of the pixel value selection. When the proposed method embeds the same capacity of secret information, the distortion of the image is better than that of the existing algorithm, and it has better visual quality.