Research On Information Hiding Algorithm Based On Homomorphic Cryptography

With the development of the Internet and cloud computing technology,people can use their powerful computing power to serve daily scientific research and creation,such as graphic and image rendering processing and model training in deep learning.In this process,it is inevitable to carry out a large amount of data transmission and calculation.This process is prone to security problems such as important information leakage,which also brings severe challenges to digital rights management protection.To solve the problem of information leakage,the commonly used method currently is to use cryptographic technology to encrypt information before transmission,in order to protect confidential information from malicious elements.In terms of digital rights management protection,information hiding algorithms are generally used to provide technical protection,for example,digital watermarking algorithms can provide functions such as right confirmation and traceability for digital rights management protection.The effective combination of cryptographic technology and information hiding technology can not only provide security for digital works,but also provide technical support for digital rights management protection,which has good practical application value.In this regard,the unique properties of homomorphic ciphers can effectively meet the above requirements.Therefore,constructing ciphertext domain information hiding algorithms based on homomorphic ciphers is of great significance for promoting the development of the Internet and cloud computing.Based on the homomorphic properties of homomorphic cryptography,this article analyzes the principles and characteristics of multiple homomorphic cryptography algorithms,and studies digital watermarking algorithms.Three information hiding algorithms are proposed.Finally,a ciphertext domain medical record book scheme is designed based on the BFV all homomorphic cryptography algorithm.The main work and achievements include the following four aspects:(1)A multi-party information hiding algorithm in the ciphertext domain was designed based on the NTRU homomorphic cryptographic algorithm.This algorithm encodes plaintext information on a polynomial truncation ring.To avoid affecting the original information during the information hiding process,it is necessary to pre design the placement positions of the original information and hidden information on the polynomial.After the original work information owner encrypts the image,the watermark information embedder embeds the watermark information into the ciphertext image through homomorphic addition operation.The experimental results show that this algorithm can embed watermark information in ciphertext images and fully restore the original work information and watermark information.(2)Based on the BFV homomorphic cryptographic algorithm,a swappable encrypted watermarking algorithm was studied and designed.This algorithm places plaintext information onto a polynomial ring.When embedding watermarks,it is possible to directly embed the watermark information onto the polynomial ring before encryption,or to use multiplicative homomorphism to embed the watermark information after encryption.When embedding watermark information,with the help of multiplicative homomorphism,users do not need to pre design the embedding position of watermark information when embedding watermark information.The order in which the watermark information is embedded will not affect the quality of the decrypted extracted information.Through experimental results,it was found that the algorithm has high efficiency and can fully recover the carrier image and watermark image.(3)A high-capacity information hiding algorithm is proposed based on the CKKS homomorphic cryptographic algorithm.Due to the CKKS homomorphic cryptographic algorithm supporting encryption and decryption operations on real and complex numbers,the original image and hidden image can be placed in the real and imaginary parts of the complex polynomial respectively when encoding plaintext information.The information hiding operation can be completed by homomorphic addition of the ciphertext of the original image and the ciphertext of the hidden image.When decrypting the recovered information,the original image and hidden image can be restored separately based on the location of the information encoding.(4)Based on the BFV homomorphic cryptographic algorithm,a ciphertext domain electronic medical record book scheme was designed and implemented.This scheme utilizes the homomorphic characteristics of the BFV homomorphic cryptographic algorithm to achieve the addition of diagnostic information in the ciphertext domain.Its characteristic is that if the patient does not need to disclose case information during consultation,the doctor can directly update the diagnostic information into the medical record book without decrypting it,greatly ensuring the security of the patient’s sensitive information.