In this paper we present a novel approach to information hiding. We investigate the possibility of embedding information using in some way the naturally pseudorandomness of some classes of cover-documents. In particular we provide algorithms for embedding any binary string in an image belonging to a particular class of images, the image mosaics. The algorithms presented allow different levels of security for the information hidden in the cover-document. We also show some techniques to reduce the amount of information the users have to secretly store.
展开▼