一种基于差值扩展大容量可逆信息隐藏新算法

A novel reversible data hiding technique using difference expansion

摘要:

在保证图像质量的前提下，为了提高可逆信息隐藏的容量，提出了一种基于差值扩展的可逆信息隐藏新算法。该方法通过差值直方图平移，将由阈值T确定的特定区间外的一部分差值平移到该区间内，同时仅对属于该特定区间的差值进行定位图标注，从而不仅增大可嵌入信息的总量，而且得到远小于现有方法的定位图。特别是在阈值T较小的情况下，效果更加明显。实验表明，该算法在保持图像低失真的前提下，与其他算法相比较，取得了较大的信息嵌入量。

Abstract:

A new reversible data hiding technique based on difference expansion was proposed. Its main idea was that some difference values which were outside the interval defined by the threshold T were shifted into the interval. A location map was created for the difference inside this interval. The proposed method could not only achieve high embedding capacity but also have smaller location map than other methods. The method was more effective especially while the threshold was small. Experiment results showed that our proposed method had better performances than other methods in terms of high embedding capacity and low distortion.