Randomness in visual cryptography

9Citations
Citations of this article
5Readers
Mendeley users who have this article in their library.
Get full text

Abstract

A visual cryptography scheme for a set P of n participants is a method to encode a secret image into n shadow images called shares each of which is given to a distinct participant. Certain qualified subsets of participants can recover the secret image, whereas forbidden subsets of participants have no information on the secret image. The shares given to participants in X P are xeroxed onto transparencies. If X is quali-fied then the participants in X can visually recover the secret image by stacking their transparencies without any cryptography knowledge and without performing any cryptographic computation. This is the first paper which analyzes the amount of randomness needed to visually share a secret image. It provides lower and upper bounds to the randomness of visual cryptography schemes. Our schemes repre- sent a dramatic improvement on the randomness of all previously known schemes.

Cite

CITATION STYLE

APA

de Bonis, A., & de Santis, A. (2000). Randomness in visual cryptography. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1770, pp. 626–638). Springer Verlag. https://doi.org/10.1007/3-540-46541-3_52

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free