Research On Reversible Data Hiding Optimization Algorithm Based On Image Pixels Histogram Shifting

Various information technologies' progress can be described as galloping due to rapid development of science and technology,they have made people's lives more convenient,but information security issues have also followed.In recent years,people's work and life have been hampered by information leakage,therefore,reversible data hiding(RDH)algorithms are proposed to protect information.RDH algorithms can hide data,extract hidden data and restore the original cover.In addition,more and more people upload personal data(such as images,audio,etc.)to the cloud for storage,and some users are worried about information leakage,so they will encrypt the data before uploading.In order to facilitate management,the cloud needs to embed some tag information in the ciphertext data,thus,the RDH technologies based on the ciphertext cover also have extremely high research value.This dissertation proposes two RDH algorithms for spatial image separately in plaintext state and ciphertext state based on histogram shifting technology.The main research achievements of this paper are as follows:(1)Reversible data hiding based on reducing invalid shifting pixels in histogram shiftingConsidering the effect of natural image texture on embedding distortion,this dissertation proposes a RDH scheme to reduce the invalid shifting pixels(ISPs)in the histogram shifting.Specifically,the fluctuation value of each pixel is calculated by a designed smoothness evaluation function,and the additional data is preferentially embedded in the prediction error with small fluctuation.Experimental verified that the proposed scheme has better visual quality and high embedding capacity compared with some related schemes.(2)Reversible data hiding of encrypted images based on Paillier public key cryptosystemAiming at an existing RDH scheme based on public key encryption system,this dissertation proposes an optimization scheme.The proposed scheme performs homomorphic operations on the images encrypted with the Paillier system to achieve histogram expansion and histogram shifting in the plaintext domain.A more efficient mapping method allows the symmetrical domain of the ciphertext domain about the X-axis to be fully utilized.Compared with some related algorithms,the proposed scheme has the characteristics of high embedding capacity,real reversibility,high security,and no preprocessing.This dissertation summarizes the research works,and points out the next research plan,as well as the potential application of RDH.