Research On Reversible Data Hiding And Encryption Algorithm For Color Image With High Visual Quality

With the development of Internet technology,people’s daily life has become more convenient.However,while the Internet brings convenience,it also brings problems such as digital image information security,copyright protection,and integrity authentication.However,data hiding is one of the main solutions to these problems.Data hiding can be divided into three categories: digital watermarking,steganography,and reversible data hiding.Among them,digital watermarking focuses on robustness,while steganography focuses on concealed transmission,and reversible data hiding is mainly used for copyright protection and integrity authentication.The main difference from the former two is that it can recover the cover image and messages losslessly.When the cover image are medical,military image,etc.,where the image needs to do integrity authentication and completely recovery,the recoverability of reversible data hiding perfectly fits the above scenario.At the same time,the security of information transmission is also crucial,and encryption is one of the main solutions.Encryption technology is mainly to encrypt the image into a ciphertext image that is far from the original image,so as to prevent it from being modified by hackers.Therefore,in order to protect private information and image security,this dissertation proposes a reversible data hiding and encryption algorithm for color images with high visual quality.The specific research content are shown in the following two points:1.Propose reversible data hiding algorithm for color images with high visual quality based on the B-R-G embedding principleDue to the difference between color images and grayscale images,first,this dissertation propose B-R-G embedding principle according to the different visual perception in three channels to improve the visual quality of the marked color images.Second,this dissertation propose double-layer LS prediction by combining the characteristics of Rhombus prediction and local LS prediction to achieve the sorting requirement and maintain sufficient accuracy of prediction.Finally,introduce a new embedding scheme by using the energy threshold and maintaining the inter-channel correlation to reduce the embedding distortion.Experiments show that the algorithm proposed in this dissertation is not only superior to existing algorithms in the visual quality of encrypted images,but also in the embedding capacity.2.Propose a reversible data hiding algorithm for color images in ciphertext domain based on the B-R-G embedding principleIn order to ensure the high visual quality of the digital image embedded with the private information,the security during the transmission process,and the lossless recovery of the image and private information.This dissertation proposes a reversible data hiding algorithm for color images in ciphertext domain based on the B-R-G embedding principle.Among them,in order to protect user privacy,this dissertation first chooses to hide the private information with RDH before encrypting because it can avoid privacy leakage in the cloud.At the same time,in order to reversibly embed the private information into the cover image and ensure the high visual quality of the embedded image.Therefore,this dissertation first uses the reversible data hiding technology based on the B-R-G embedding principle introduced in the first part to embed the private information;in addition,in order to ensure that the embedded image in the transmission process is not too much visually concerned,this dissertation chooses to encrypt the embedded image in plaintext,that is,encrypt the embedded image into an image similar to the target image.Experiments show that the visual quality of the encrypted image generated by the reversible information hiding method in this paper is high,and the encrypted image is also similar to the target image,and far from the original image.It not only achieves the encryption effect,but also turns the encrypted image into a plaintext image.