Region Incrementing Visual Secret Sharing Scheme
Based on Random Grids
Guan-Shi Zhong and Jian-Jun Wang
Department of Electronic Engineering
Fudan University
Shanghai, China
E-mail: zhonggshi@163.com, wangjj@fudan.edu.cn
Abstract—A region incrementing visual secret sharing (RIVSS)
scheme can incrementally reveal secrets with different secrecy
levels. Recently, Wang proposed a 2-level incrementing visual
cryptography using random grids, which can incrementally
reveal the secrets on an image without pixel expansion.
However, Wang’s scheme can only deal with 2-level RIVSS. In
this paper, a novel region incrementing visual secret sharing
(RGRIVSS) scheme based on (k, n)-threshold random grid is
proposed. The proposed RGRIVSS scheme can deal with more
than 2-level region incrementing VSS. The experimental results
demonstrate the feasibility of the proposed scheme.
I. INTRODUCTION
Visual cryptography (VC) was first proposed by Naor and
Shamir [1] in 1994. It is a kind of visual secret sharing (VSS)
technique which is free of computation during the decoding
process. In the (k, n) VC scheme (2 ≤ k ≤ n), the secret image
is encoded into n transparencies, called shares, only if at least
k shares are superimposed together can reveal the secret
image, but no information can be obtained by any less than k
shares. The most attractive feature of VC is that the decryption
process relies on human visual system without any
computation and sophisticated knowledge of cryptography.
Such interesting property attracts much attention from many
researchers. The existing literatures focus on reducing the
pixel expansion [2], using halftone technique for gray and
color images [3, 4], etc.
Recently, a region incrementing visual secret sharing
(RIVSS) scheme is proposed by Wang [5]. Essentially, the
conventional VC schemes focus on processing the content of
an image as a single secret while RIVSS scheme considers the
secret image with multiple regions, where each region has its
own secrecy level. In the decoding process, the more shares
are superimposed, the more secrets with higher secrecy level
will be revealed. However, Wang only gave the encoding
basis matrices of n = 3, 4 and 5, and no construction method
was discussed. Therefore, Yang [6] proposed a general (k, n)-
RIVSS scheme, where k and n are any integers. Almost at the
same time, Shyu and Jiang [7] proposed an efficient
construction for RIVSS which is based on integer linear
program aiming at the minimization of the pixel expansion.
The aforementioned schemes suffer the drawback of pixel
expansion problem and basic matrices designing. To solve this
problem, Wang et al. [8] proposed an incrementing visual
cryptography using random grids (IVCRG). However, only the
construction of 2-level IVCRG was given by Wang.
In this paper, based on (k, n)-threshold random grids VSS
scheme, a novel n-level region incrementing visual secret
sharing (RGRIVSS) scheme is proposed.
II. P
ROPOSED METHOD
A. Region Incrementing Visual Secret Sharing Scheme
In n-level region incrementing VSS scheme, the secret
image is firstly divided into n disjoin regions L
0
, L
1
, … , L
n
where L
0
represents the background, and L
i
are treated as the
foreground with the i-th secrecy level for 1 ≤ i ≤ n. Then the
secret image is encoded into n+1 shares. Superimposing any t
shares can reveal the region secret with t-1 secrecy level, but
cannot reveal any region secrets with t, t+1, … , n secrecy
levels (2 ≤ t ≤ n).
By using random grids, Wang et al [8] proposed the
necessary conditions for the light transmission of the n levels
of secrets and the background when t shares are superimposed.
Let
0
t
c and
t
i
c represent the light transmission of the
background L
0
and the i-th level secret L
i
respectively when t
(1 ≤ t ≤ n+1) shares are stacked together. The conditions can
be expressed as
; , 0,1,..., 1
tt
ij
ccijij t≠≠ = − (1)
0
, 1,...,
tt
i
ccitt n==+ (2)
It states that the more shares are gathered, the more secret
can be revealed incrementally.
B.
(k, n)-threshold VSS scheme Based on Random Grids
The (k, n)-threshold visual secret sharing scheme based on
random grids (RGVSS) comprises four phases:
•
Step 1. Encode a secret pixel S(i, j) into k sub-pixels
r
1
(i, j), …, r
k
(i, j) by using (k, k)-RGVSS scheme [9].
This research was partly supported by National Natural Science Foundation
of China under Grant No. 61170207. Corresponding author: Jian-Jun Wang.
978-1-4673-5762-3/13/$31.00 ©2013 IEEE 2351