基于图像块相关性分类的加密域可逆数据隐藏

Reversible Data Hiding in Encrypted Domain Based on Correlation Classification of Image Blocks

考虑到不同图像块中像素的重要位(高位)具有不同的相关性,本文提出了一种基于图像块高位相关性分类的加密域可逆数据隐藏算法.在数据嵌入前,对各图像块的高位进行相关性划分,然后根据相关性强弱对加密图像的相应像素执行不同程度的低位(不重要位)取反,以达到减小嵌入失真和提高数据提取正确率的目的.同现有的两个主要方法相比,大量的实验结果表明该算法在保证解密后图像具有相同峰值信噪比的情况下,能有效减小数据提取和图像恢复的错误率.

This paper proposes a new reversible data hiding in encrypted domain by considering the correlation of MSBs（most significant bits）of image blocks.For an information bit embedding in a block,its MSBs＇ correlation is computed for deciding the number of LSBs（Least Significant Bits）to be flipped.In such a way,data recovery rate can be improved for the same embedding distortion by comparing the two existing states of the art works.Experiment results have proven the validity of the proposed method.