Reversible Data Hiding In Images Using Histogram Processing

As an important branch of data hiding,image reversible data hiding(RDH)hides data into a digital media reversibly,which means data receiver can retrieve the original media exactly after data extraction.The technology is widely applied on cloud services and cloud storage.Histogram processing is an important technique widely applied on data hiding.This paper reviews several typical and state-of-the-art image RDH methods.This paper further proposes two novel RDH schemes.The main contribution of this paper can be concluded as follows:1)A novel framework of RDH with contrast enhancement(RDH-CE)using histogram shifting is proposed: First,this paper proposes an RDH scheme based on histogram vacancy reservation.The scheme first merges the least significant bins to reserve spare bins and then embed additional data by a histogram shifting approach using arithmetic encoding.During histogram shifting,an algorithm is designed for the construction of the histogram of transfer matrix,which aims at maximizing the entropy of the histogram.After embedding,the marked image generally has a better image quality than the original image.In order to enlarge embedding capacity,this paper further proposes to concatenate the above-mentioned scheme with an MSE-based embedding.On the recipient side,the additional data can be extracted without error.Also,the cover image can be recovered exactly.The proposed method generates marked images with good quality with the sense of structural similarity.Comparing with existing RDH-CE approaches,the proposed method can achieve a better embedding payload.2)An interpolation-based reversible data hiding scheme in encrypted images(RDH-EI)is proposed: The scheme aims at avoiding image information leakage.Before image encryption,the embedding space is first spared by conducting difference expansion of interpolation-errors of non-sample pixels.Auxiliary data generated by difference expansion is then embedded into sample pixels.Then,sample pixels are encrypted using a stream cipher.And non-sample pixels are encrypted using a scrambling cipher,which preserves the histogram of interpolation-errors.The data hider embeds additional data into the spared interpolation-errors using histogram modification techniques.Finally,authorized receiver can extract embedding data exactly and recover the original image perfectly.The proposed scheme is proved to be more effective and provides better security compared to some start-of-the-art methods.Image contours can be well hided.Also,the embedding capacity can be adjusted by the data hider,which caters to user's requirement.