抵抗唯密文攻击的可分离加密域可逆信息隐藏

Separable Reversible Data Hiding in Encrypted Image Against Ciphertext-Only Attack

为提高密文图像安全性和算法的可逆性,提出一种抵抗唯密文攻击的可分离加密域可逆信息隐藏算法.图像拥有者将图像不重叠地分块后,采用不同密钥分别异或加密块内像素的高3位和低5位,再进行块置乱得到密文图像.信息隐藏者利用像素波动性和无损压缩技术获得较大的嵌入容量,以位翻转方式隐藏附加信息,并通过记录像素波动性实现算法的完全可逆.图像接收者既可无差错提取信息,又可无损恢复原始图像.选取Uncompressedcolor image database图像集中的图像进行实验,结果表明,在数量为20,50,70,100的密文图像集上实施唯密文攻击所得流密码正确率均为50%左右,有效地提高了算法抵抗唯密文攻击的能力;与3种同类算法相比,该算法的隐藏容量分别提升了145.69%,148.21%,240.85%,且得到的解密图像质量均提高约5 dB.

To improve the security and reversibility,a reversible data hiding in encrypted image capable of ciphertext-only attack(COA)is proposed.The image owner segments the original image into a number of non-overlapping blocks and the encrypted image is generated by combining block scrambling and the designed XOR encryption which three MSBs and five LSBs of pixels in the block are encrypted respectively with different key stream.And then the data hider uses the pixel fluctuation value and lossless compression to gain extra space and embeds the additional information into pixels by bit flipping.To achieve full reversibility,the fluctuation value of the pixels is recorded in marked image.At last,the receiver extracts the embedded data correctly and recovery original image.The proposed algorithm was evaluated on uncompressed color image database.Experimental results show that the correct rate of stream cipher obtained by COA on encrypted image sets of 20,50,70,100 is about 50%,which can effectively resist COA.Compared with three similar algorithms,the embedding capacity is improved respectively by 145.69%,148.21%and 240.85%,and the quality of decrypted image is improved about 5 dB.