Abstract
The (t, n) visual cryptography (VC) is a secret sharing scheme of decomposing a secret image into n transparencies, and the stacking of any t out of n transparencies reveals the secret content. The perfect security condition of VC scheme requires the strict requirement where any t-1 or fewer transparencies cannot extract any information about the secret. For n approaching infinity, previous studies consider the scenario where the probabilistic model is a pixel-to-pixel scheme that encodes each secret pixel to a corresponding pixel in each transparency. In this paper, we extend the pixel-to-pixel scheme to pixel-to-block scheme for the cases t=2 and 3. Given a secret image, the proposed VC scheme generates a transparency through coding each secret image pixel to a m-pixels shadow block on the corresponding position of the transparency. Experiments show that the stacking results reveal better visual quality than the probabilistic model scheme. © 2012 APSIPA.