Visual Cryptography Schemes with Multiple Secrets and Visual Key Derivation

Abstract

Visual Cryptography was introduced in 1995 by Naor and Shamir to recover a secret image by overlapping two or more images. This scheme is secure and easy to implement and can be extended to a set of participants in order to recover the same secret image. In this thesis, we propose two new schemes that use a shared image to encode a different secret image for each participant. In the first scheme, we generate a visual cyphertext from a shared key for each participant. Since the shared key is computed a priori, we can add a new participant at any given time and generate her cyphertext, ensuring scalability. In the second scheme, we make the approach more practical by applying a visual key derivation function in order to let each participant derive her visual cyphertext from a different password for each participant. This is a joint work with colleague Tommaso Moretto, who has developed implementations and performed practical experiments on these new schemes applied to barcode confidentially.