Block Halftoning for Size-Invariant Visual Cryptography Based on Two-Dimensional Lattices

Abstract

Visual cryptography (VC) encodes a secret image into multiple transparencies. The secret image can be revealed by stacking a subset of these transparencies. However, the pixel expansion problem is a critical issue for VCs as a pixel in the secret image is encoded into multiple pixels. To solve this problem, many size-invariant VCs have been proposed. But there are some drawbacks in the existing size-invariant VCs. i) The input image is a binary image rather than a grayscale image. ii) Some schemes are not a ($k, n$) - VC. iii) The quality of recovered image is poor. Based on two-dimensional lattices, a block halftone technology for a size-invariant VC is proposed. In particular, the proposed block error diffusion takes a grayscale image as the input. It is adaptable to any pixel expansion. The experiments show that the proposed halftone technique has a better visual quality for the stacking results than prior common schemes, such as probabilistic schemes.