摘要

In a (k, n)-VSS (visual secret sharing, 2 <= k <= n) scheme, a secret image is encrypted into n shares that are distributed among n participants. Reducing pixel expansion and improving the quality of recovered images to enhance the usefulness are still major issues of VSS scheme constructions, especially for large k and n. Moreover, all existing constructions of VSS schemes have to be designed individually; developing a generic construction or method for VSS schemes is still an open question. In this study, an optimization technology is used to develop a generic probabilistic VSS (GProb-VSS) (k, n)-scheme without pixel expansion for binary secret images. First, we define a mathematical optimization model to maximize the contrast in recovered images under security constraints. Then, we develop a simulated-annealing-based algorithm for solving the difficult problem. Furthermore, we try to improve the display quality by relaxing the relative parameter. Experimental results show that the proposed generic approach significantly outperforms the previous methods in terms of the recovered image quality and pixel expansion.

  • 出版日期2011-9