Contrast Optimal Threshold Visual Cryptography Schemes

From MaRDI portal





The Threshold Visual Cryptography Schemes (VCS) were introduced by \textit{M. Naor} and \textit{A. Shamir} [Lect. Notes Comput. Sci. 950, 1--12 (1995; Zbl 0878.94048)] and have been widely studied since then. A \((k,n)\)-threshold VCS is a method of sharing a secret (black and white) image among a set of \(n\) participants, giving to each of them a \textit{share} of the image in such a way that any set of \(k\) or more participants can recover the image (without any cryptographic computation: only copying their shares into transparencies and stacking them) but any set of less than \(k\) participants has no information about the secret image. For doing so, each pixel is encoded in \(n\) versions, each of them a collection of \(m\) black and white subpixels. The number \(m\) is called the pixel expansion. Other authors have extended the scope of the Naor-Shamir model: [\textit{G. Ateniese, C. Blundo, A. de Santis} and \textit{D. R. Stinson}, VCS for general access structures, Inf. Comput. 129, 86--106 (1996)]; [\textit{E. R. Verheul} and \textit{H. C. A. van Tilborg}, Des. Codes Cryptography 11, 179--196 (1997; Zbl 0878.94049)] shares images that conceal the existence of a secret image (they look like innocent recognizable images: a dog, a house); [\textit{G. Ateniese, C. Blundo, A. de Santis} and \textit{D. R. Stinson}, Theor. Comput. Sci. 250, 143--161 (2001; Zbl 0952.68056)], etc. In the present paper the authors analyze the contrast of a VCS (the contrast is a value \(\alpha\geq 1/m\)\, that measures the difference between a black and a white pixel in the reconstructed image) and the maximum possible value of the contrast (contrast optimal) for any \(k,n\). The first two sections are introductory. In section 3 the authors define a canonical form for a \((k,n)\)-threshold VCS and they completely characterize contrast optimal \((n-1,n)\) VCS in that canonical form (subsection 4.1). For \(n\geq 4\) they prove an upper bound on the contrast of any \((3,n)\)-threshold VCS and provide a \((3,n)\) contrast optimal VCS (subsection 4.2). Finally for \(k=4\) (section 5) and \(k=5\) (section 6) they provide families of \((k,n)\)-threshold VCS achieving various values of contrast and pixel expansion and determine schemes in those families with the best contrast (in the paper's conclusions they conjecture that those schemes have optimal contrast).




Cited in
(34)








This page was built for publication: Contrast Optimal Threshold Visual Cryptography Schemes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4443082)