本論文提出一個兩層式結構的(t, s, k, n)關鍵秘密影像分享門檻技術，當中有s個關鍵分享影像和n-s個非關鍵分享影像，且還原回秘密影像時，需收集t個關鍵分享影像和k個分享影像，其中分享影像包括關鍵分享影像和非關鍵分享影像。在此技術中定義一個參數m，m=1時，將獲得最小的總分享影像的影像大小及關鍵分享影像的影像大小接近於非關鍵分享影像的影像大小；m=t時，將獲得關鍵分享影像的影像大小和非關鍵分享影像的影像大小為2倍關係，並且能凸顯關鍵分享影像的重要性，此外，本論文克服前人Chen的方法[3]所存在之不滿足分享門檻問題，不會造成還原時以滿足單一門檻能獲得部份秘密影像資訊的缺點，實驗結果顯示，本文方法可歸納出兩種最佳分享率的分享結果。 This paper presents a two-layered structure for optimally sharing a secret image among s essential and n-s non-essential shared shadows using the (t, s, k, n) essential thresholds, that t essential shared shadows and totally k shared shadows are needed to recover the secret image. The presented two-layered structure includes one user-defined parameter m to determine different kinds of optimal results. m=1 leads to minimum size of total shared shadows (ST) and size of an essential shared shadow is close to size of a non-essential shared shadow. On the other hand, m=t leads to size of an essential shared shadow being twice of size of a non-essential shared shadow to signify the importance of an essential shared shadow. Moreover, the proposed structure overcomes the threshold fulfillment problem in Chen’s scheme [13]. Theoretical analyses and experimental results show that the proposed scheme exhibits secure with optimal sharing ratios among related works.
