階層式且無像素擴展之視覺多秘密分享機制

Abstract

傳統的視覺秘密分享機制將一個秘密影像分解成數張分享影像，以提供足夠的安全性。現今的技術，已經可以將多個秘密影像藏於相同集合中的分享影像，只要將分享影像旋轉到某個角度即可看見另一個秘密影像。然而這些方法被侷限在(2, 2)或者(3, 3)-視覺秘密分享機制，而且這些分享影像的大小也被擴大。 在這篇論文中，我們提出一個改善的視覺多秘密分享機制，使用階層式的概念來管理這些分享影像。只要將某個集合中的分享影像重疊即可得到第一個秘密，第二個部份秘密的取得需要藉由某些分享影像的位移，再將其與其他的分享影像重疊，即可獲得。最後再將所有的部份秘密蒐集合併在一起，便可獲得第二個完整的秘密。我們的方法可以被延伸至(k, k)-視覺秘密分享機制，而且分享影像的大小將會與原始的秘密影像大小相同。

Traditional visual secret sharing scheme has provided sufficient security for a secret image by breaking it into share images (called shares). In current technology, multiple secret images can be hidden in the same set of shares after one of the shares is rotated at different degrees. However, these schemes are limited to (2, 2) or (3, 3)-visual secret sharing. Furthermore, the sizes of the shares will be expanded. In this thesis, we propose an improved visual multi-secret sharing scheme that adopts a hierarchical concept to manage the shares. The first secret can be obtained by stacking some subset of shares, and the second fragmented secret can be obtained by shifting some shares in the same set and stacking them together. Then, we merge all the fragmented secrets to obtain the whole information about the second secret. Our scheme can be generalized to (k, k)-visual secret sharing, and the sizes of shares remain the same as the secret image.