Two-Decoding-Option Image Sharing Method - Visual Cryptography and Secret Image Sharing - page 468

(a)G 1 (b)G 2 (c)G 1 +G 2

FIGURE 17.5
The proposed (2, 2)-TiOISSS using base matrices with h = 1;l = 0, and
m = 3: (a) and (b) two 512 512 gray-and-white shadows (c) the previewed
image.

according to the gray pixels in P 1 and P 2 . In Figure 17.4(b) and Figure 17.4(c)
are two gray-and-white shadows G 1 and G 2 of the size mjI 0 j = jIj (512512-
pixels). Figure 17.4(d) is the previewed image by stacking G 1 and G 2 without
computation. In the second phase of decoding, we can obtain P 1 and P 2 from
the gray pixels of G 1 and G 2 , and then reconstruct the gray-level Lena in
Figure 17.4(a) by I = P -1 (P 1 ;P 2 ).
Case (2) B 1 =

110
101

110
110

, B 0 = :

Since m = 3 and g = 2, we need a halftone image I 0 of the size is jI 0 j =
jIj=(k g) = jIj=4. Output two shadows G i = G(I 0 ), i = 1, 2, by using
(2, 2, 3, 2)-GVCS, and the values of gray subpixels are chosen according to
the gray pixels in P 1 and P 2 . In Figure 17.5(a) and Figure 17.5(b) are two
gray-and-white shadows G 1 and G 2 of the size mjI 0 j = 0:75jIj (512384-
pixels). Figure 17.5(c) is the stacked result by stacking G 1 and G 2 . By the
same approach in Case (1), we can also reconstruct the original gray-level
Lena I.

2
All the above schemes can visually reveal the secret by simply stacking
shadows in the preview phase. The original gray-level secret image can be
perfectly reconstructed in the perfect-reconstruction phase. Of two (2, 2)-
TiOISSSs in Example 4, Figure 17.4(d) possess the better resolution of the
previewed result, but Figure 17.5(c) has a lesser shadow size of 512 384
pixels.
Because our TiOISSS uses PISSS in Phase 2 for decoding, we can use
the compression approach in [9] to hide the compressed image into shadows.