Abstract
In recent years, various visual secret sharing schemes have been proposed to protect digital images from malicious behavior with comprehensive properties such as (k, n) threshold, simple recovery, no pixel expansion, meaningful shadow images, alternative order of shadow images, and avoiding codebook design. However, to the best of our knowledge, no previous secret sharing scheme achieves almost all the above properties. In this paper, a visual secret image sharing threshold scheme is proposed based on random grids and error diffusion. It has several performances to other schemes such as (k, n) threshold, simple computation (stacking recovery), alternative order of shadow images in recovery, no codebook design, and avoiding the pixel expansion problem. In addition, it has meaningful shadow images, at the price of possible degrading of images' quality. Experiments are conducted to show the security and efficiency of the proposed scheme. Comparisons with previous approaches show the performances of the proposed scheme.