Reversible Data Hiding Algorithms In Encrypted Images Based On Differential Compression And Homomorphic Encryption

With the popularity of cloud computing,mobile Internet and smart terminals,public cloud platform has become ideal storage and computing service provider,and more and more people prefer to store multimedia data in the cloud.To protect personal privacy and ensure data security,users usually want to encrypt multimedia data before uploading it to the cloud,but this is inconvenient for further data processing,such as data compression,data storage and information retrieval.To this end,researchers design encrypted domain signal processing technology,which includes outsourcing computing,ciphertext compression,ciphertext retrieval,and ciphertext data hiding.In addition,how to realize annotation,storage query and usage management of encrypted images under the premise of protecting the privacy of image owner has become the research hotspot.This paper focuses on the frontier problems of encrypted images data hiding and carries out in-depth research on encrypted image reversible data hiding technology with differential compression,homomorphic encryption and histogram shifting.Main research works are summarized as follows:1.Reversible data hiding with differential compression in encrypted images is designed.There are two key contributions in this algorithm:(1)An efficient block-based image encryption scheme is proposed,which can transfer spatial correlation between neighboring pixels of plaintext image into encrypted image;(2)An image compression method called differential compression is proposed,which can efficiently compress encrypted image by exploiting pixel correlation and vacate a large room for data embedding.Specifically,in the stage of data embedding,encrypted image is firstly generated using the proposed block-based image encryption scheme.After that block difference compression and block classification are conducted to determine available image blocks.And then structure information and secret data are embedded block by block,and finally auxiliary information is also embedded to finish the entire data hiding process.In the stage of data extracting,auxiliary information is extracted firstly,then secret data is extracted and encrypted image is recovered block by block.Experimental results show the proposed algorithm has a large embedding capacity and the features satisfy the requirement of large-capacity reversible data hiding in encrypted images.2.Reversible data hiding with shifting block histogram of pixel differences in encrypted images is established.There are two key contributions in this algorithm:(1)An image encryption method maintaining the homomorphic addition characteristics is proposed;(2)An embedding method based on block difference histogram shifting is designed,which can efficiently implement data embedding.Specifically,in the stage of data embedding,the proposed image encryption method is performed to generate encrypted image.And then the secret data and auxiliary information is embedded with histogram shifting after block pixel difference histogram constructing.In the stage of data extracting,auxiliary information is extracted first,then conducting histogram re-shifting to extract secret data and recover encrypted image.Experimental results show that the proposed algorithm achieves higher embedding rate,better resulting image quality and greater computational efficiency than compared methods.