Study Of Digital Watermarking Algorithm Based On Homomorphic Public Key Cryptosystem

The application of cloud computing has become more and more popular in recent years,which has provided services for people as well as caused many problems of data security,how to effectively protect the data in cloud is a hot research of cloud computing security.The digital watermarking based on homomorphic public key cryptosystem provides a solution for privacy and security protection of the digital media in cloud,which has good research significance and practical value.Therefore,two digital watermarking algorithms based on homomorphic public key cryptosystem are proposed in this paper:1)A robust reversible watermarking algorithm in homomorphic encrypted domain based on histogram shifting of statistic values of encrypted blocks is proposed.The proposed algorithm adopts a strategy of image-blocks encryption and designs a method to retrieve statistic values of the encrypted blocks,after that,watermark can be embedded into the encrypted image by shifting the histogram of the statistic values by employing the homomorphic and probabilistic property of the cryptosystem?modular inverse method and looking for a mapping table.The embedded watermark is robust to those common image processing operations(JPEG/JPEG2000 Compression and Gaussian Noise)to some extent.2)A reversible watermarking algorithm in homomorphic encrypted domain based on expanding of statistic values of encrypted blocks is proposed.The proposed algorithm also adopts the strategy of image-blocks encryption and the method to retrieve statistic values of encrypted blocks proposed in the last algorithm,so that watermark can be embedded into the encrypted image by expanding the statistic values.The proposed algorithm has feature of high embedding capacity and low embedding distortion.Compared with the existing reversible watermarking algorithm in homomorphic encryption domain,the proposed algorithm still has good quality of the watermarked and decrypted image at high embedding rate,and effective maximum embedding capacity has been improved to a certain degree.