Interactive Screen-Shooting Resilient Ciphertext Watermarking Protocol Between the Cloud and the User

Xiaojuan DONG,Weiming ZHANG,Han FANG,Nenghai YU
DOI: https://doi.org/10.1360/ssi-2021-0213
2022-01-01
Abstract:The current cloud storage services generally launch the function of browsing files, which brings greatconvenience for users to steal information by shooting files on the screen, and also increases the risk of information leakage. As a result, users taking photos to steal information poses a potential security risk for cloud storage services.The existing method is for the cloud to embed a screen-shooting resilient watermark associated with the user's identityin an image before sending it to the user. Then the cloud can detect the user's watermark from the captured image and discover the user's identity. Both the cloud and the user have the image embedded with the user's watermark, so the leaked images cannot be confirmed to be from the cloud or from the user. To eliminate the ambiguity of the leakage source, ciphertext watermark is utilized, that is, the user's ciphertext watermark is embedded into a plaintext image in the cloud to generate the ciphertext image containing the user's watermark, which can only be decrypted by the user to obtain theplaintext image. Once this image is secretly shot and leaked, it must flow out from the user. However, the current screen-shooting resilient watermarking algorithm does not support the above functions and only supports the embedding of the plaintext watermark. This paper proposes a screen-shooting resilient ciphertext watermarkingprotocol between the cloud and users, which aims to protect the copyright users' ownership of their images and prevent other userswho use the images from stealing images by taking pictures secretly. This paper verifies the security and effectiveness of the proposed protocol through theoretical analysis and simulationexperiments.
What problem does this paper attempt to address?