POISIDD: privacy-preserving outsourced image sharing scheme with illegal distributor detection in cloud computing

• Published in: Multimedia tools and applications Vol. 81; no. 3; pp. 3693 - 3714
• Main Authors: Deng, Tianpeng, Li, Xuan, Xiong, Jinbo, Wu, Ying
• Format: Journal Article
• Language: English
• Published: New York Springer US 2022; Springer Nature B.V
• Subjects: Authentication; Cloud computing; Computer Communication Networks; Computer Science; Data Structures and Information Theory; Discrete cosine transform; Distributors; Embedding; Floating point arithmetic; Multimedia Information Systems; Outsourcing; Privacy; Special Purpose and Application-Based Systems; Data hiding; Illegal distribution detection; Image sharing; Homomorphic encryption; Discrete cosine transform
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-021-11737-8

Privacy-preserving outsourced image sharing schemes can help relieve the local storage of image owners and protect images’ privacy by sending encrypted images to receivers, but they cannot prevent the decrypted image from being illegally distributed by illegal distributors. To cope with this issue, we propose a privacy-preserving image sharing scheme with illegal distributor detection (POISIDD). POISIDD embeds the authentication information of image receiver in the encrypted form based on discrete cosine transform (DCT) and privacy-preserving outsourced calculation on floating point numbers (POCF) . Moreover, in order to prevent all operations from being carried out by the single cloud which may lead to privacy disclosure, we adopt two platforms (Storage Center and Authentication Center) to achieve the embedding of authentication information( AI ) by the communications between them. Furthermore, in order to better defend against attackers, we generate random vectors based on the original AI and use them instead of the AI as the embedded data. The aim of identifying illegal distributor will be achieved with the help of Authentication Center. The scheme is secure under the defined attack model, and its robustness under different attacks has been shown by experiments.