- 學位論文
基於差值擴張之可逆機密影像分享技術
Reversible secret image sharing scheme based on difference expansion
摘要
(t, n)門檻式機密影像分享技術是將機密影像分成n張無意義的分享影像並分配給n位參與者，其中任何超過t位參與者的群組，都能利用他們的分享影像還原機密影像。反之，則不能還原機密影像。同時，為降低分享影像的顯著性，可進一步藉由偽裝學(Steganography)技術，將分享影像藏入有意義的掩護影像中形成一張偽裝影像，來避免攻擊者的注意力與提供管理上的便利性。本文提出結合偽裝學技術的可逆機密影像分享技術，讓參與者在還原過程中，除了可以從偽裝影像中擷取機密影像外，還可以將偽裝影像回復成掩護影像。本文採用m進位下之差值擴張方法配合m進位的機密影像分享技術，可以得到高嵌入量及高影像品質之無失真影像秘密分享技術。實驗結果顯示，本論文所提出的方法，偽裝影像有良好的品質且掩護影像有很高的嵌入量。
並列摘要
Conventional (t, n) secret image sharing scheme shares a secret image to shared images and secret keys. Collecting t shared images corresponding with their secret keys recovers the secret image. However, noise-like shared images always attract attackers attention. Therefore, embedding shared images to meaningful cover images to form stego images can efficiently reduce attackers attention. This paper presents a different-expansion based reversible secret image sharing technique to let participants restore the secret image and cover image in recovering procedure. The proposed scheme can further raise the embedded capacity when the cover image does not need to be perfectly reconstructed. Experimental results show each stego image have good image quality and high embedded capacity.