Study On Reversible Data Hiding In Encrypted Images Based On Homomorphic Encryption

With the progress of technology and the arrival of the Internet era, network has penetrated into all aspects of people’s work and life. Network servers as a platform for the free exchange, and people can almost copy and modify all kinds of information in unlimited ways, which will lead to the fact that secret information and trade secrets may be stolen and tampered and copyright disputes occur, and these will be harm to social stability. Therefore, how to guarantee the security of network information has become the focus. In the practical application, we can combine the encryption technology and reversible data hiding technology to further strengthen the security of information.Most of the existing literatures use the XOR operation to realize image encryption, and the reversible data hiding algorithm based on the homomorphic encryption is proposed in this paper, which applies homomorphic encryption to the reversible data hiding in encrypted images. The main contents are as follows:(1) A reversible data hiding algorithm based on the homomorphic encryption is proposed for encrypted images. In this algorithm, firstly, the original image is encrypted by using the homomorphic encryption algorithm. Secondly, the data-hider segments the encrypted image into some nonoverlapping image blocks and divides the pixels of each block into two sets, and then modifies the pixels of one set by image fusion to embed the secret data into the encrypted image. Finally, the embedding algorithm is realized by add operation because the encryption algorithm meets with additive homomorphism, and then the receiver can decrypt the image containing the secret data, and the data hiding process is same as the operation for the original image. Thus the receiver can use extraction algorithm according to pixels correlation to extract the secret data and obtain the recovered image. Finally, the recovery performance of the proposed algorithm is analyzed by the simulation results.(2) A reversible data hiding algorithm based on content features is proposed for encrypted images. Firstly, the original image is preprocessed before encryption. The image is segmented into some nonoverlapping blocks, every block is classified fluctuant, middle or smooth area, and different operations are performed for the different area. Secondly, preprocessed image is encrypted and the data-hider embeds the secret data into the encrypted image. Finally, the receiver can use extraction algorithm to recover the preprocessed image, and then recover the original image after inverse preprocessing. Compared with the algorithm proposed in the third chapter, the difference between the pixels in the fluctuant region and the middle region becomes smaller, the correlation becomes better, and the image becomes smoother, which reduces the error rate of the extracted data and gets better recovered image.