Secure Image Deduplication in Cloud Storage

Abstract

Part 8: Multimedia SecurityInternational audienceWith the great development of cloud computing in recent years, the explosive increasing of image data, the mass of information storage, and the application demands for high availability of data, network backup is facing an unprecedented challenge. Image deduplication technology is proposed to reduce the storage space and costs. To protect the confidentiality of the image, the notion of convergent encryption has been proposed. In the deduplication system, the image will be encrypted/ decrypted with a convergent encryption keywhich is derived by computing the hash value of the image content. It means that identical image copies will generate the same ciphertext, which used to check the duplicate image copy. Security analysis makes sure that this system is secure

Survey of image deduplication for cloud storage

“…Image compression followed by encryption and hashing in sequence reduces the computational overhead, resources, and metadata to be stored. Authors of [12] discuss how cloud services have had an immense improvement in this year in terms of Secure Image Deduplication. Due to this great development in the services, many people have started storing data, which may also be redundant.…”

Secure Single-Server Nearly-Identical Image Deduplication

“…Note that some practical attacks are not prevented by this ideal functionality, most notably that adversaries are able to learn in a quantifiable way how similar their uploaded images are to those of another user. This is a common problem in secure deduplication schemes, and to the best of our knowledge there is no consensus in the community of how to address this concern [8], [16]. We address some of these attacks with practical safeguards discussed in Section VI.…”

“…When the server notifies two users to share encryption keys, it then also gives them a one-time-use token pair that the users can use to authenticate themselves to one another without revealing their identities. A common assumption in image deduplication is that the server must be able to know which users own which images in order to identify duplicates between different users, so we follow the precedent set by [14], [16], [27], and assume brute-force server inference attacks to be outside our threat model.…”