Research On Joint Coding And Key Technologies Of Image Encryption And Information Hiding

With the rapid development of network technology and cloud storage technology,people are increasingly inclined to store digital multimedia products such as digital images,video,audio and text in major storage software platforms.As the stored multimedia information involves the user’s personal privacy,any unauthorised operation may cause irreparable damage to the user.As an important technology for privacy protection,image encryption and information hiding technology have always received extensive attention.Joint image encryption and information hiding coding can play a dual role in protecting privacy data,but encryption technology will destroy the pixel correlation of plaintext image,and these correlations are the premise of information embedding,so it is very difficult to embed information on encrypted images.Performing preprocessing operations before encryption can better utilize the redundancy of natural images,which requires the image owner to complete a large number of complex preprocessing operations,resulting in additional computational burden;and,the image owner needs to be in close contact with the information hider and pass most of the auxiliary information such as preprocessing operations to the information hider,which has certain limitations in practical applications.Based on the above analysis,joint image encryption and information hiding coding still faces difficult and urgent problems.In this paper,based on related research technologies at home and abroad,from the perspective of digital image privacy protection,we mainly focus on the image block encryption security,robustness and transmission security issues that have yet to be solved,and the completed work includes the following aspects:(1)To address the problem that image block encryption in reversible data hiding in encrypted images(RDHEI)may lead to poor security,a novel RDHEI method based on permutation ordered binary number(POB)is proposed.The existing methods usually adopt block encryption to reserve a certain redundant space in the image encryption process,but this may leave attack clues for attackers.Therefore,in the embedding process,the POB number system is used to destroy the redundancy within the image block,and the "encrypted pixel value->binary sequence->POB value" operation frees up space for embedding.The embedding process compresses and re-encrypts the encrypted pixels and compensates for the security caused by the retention of correlation within the block.Experiments verify that the proposed method not only guarantees security,but also achieves reversibility when embedding higher capacities.(2)To address the robustness problem in image authentication,an image authentication method based on double random phase encoding(DRPE)is proposed.In the watermark information generation stage,the plain image is converted into an amplitude image and a phase image by DRPE operation,and the phase image is quantized to generate the watermark information to be embedded.To improve the robustness of the image authentication process,the singular value decomposition(SVD)method is used to embed the watermark information into the scrambled coefficients.On this basis,the watermarked image is post-processed using a diffusion operation to generate the final transmitted image.The experimental results verify that the proposed method can successfully achieve image authentication even after suffering a series of signal processing attacks.(3)To address the problem that the noise-like appearance of ciphertext can easily lead to attacks,a meaningful image encryption algorithm based on compressive sensing(CS)is proposed.The algorithm uses information hiding technology to embed the secret information into the carrier image,and obtains a meaningful encrypted image with a similar appearance to the carrier image,which ensures the transmission security of the ciphertext content to a certain extent.By introducing CS technology,the plaintext image is compressed and encrypted into a smaller secret image,and the secret image is embedded into the carrier image of the same size as the plaintext image by using the smooth function embedding method to generate a meaningful encrypted image.The experimental results verify that the proposed method is effective in improving the visual security of encrypted images and the quality of decrypted images.