標題: 多張機密影像的視覺加密技術
On Visual Cryptography of Multiple Secret Images
作者: 李豐政
Li, Fong-Jheng
林志青
Lin, Ja-Chen
多媒體工程研究所
關鍵字: 多機密影像;視覺密碼;投影片;重疊方式;Multiple secret images;Visual cryptography;Transparency;Stacking manner
公開日期: 2013
摘要: 本論文提供三種不同的多機密影像之視覺加密方法,分別為:(2,2)型、有條件的(2,3)型、對折型。在(2,2)型中,其中又區分為二種方法:1)多重疊合分享法,將十二張機密影像編碼成二張投影片。當拿到這二張投影片時,藉由多種不同的重疊方式(左右翻轉、旋轉、平移),可分別還原這十二張機密影像。2)高資訊隱藏量分享法,將四張機密影像編碼成二張投影片。當取得這二張投影片時,可由不同的重疊方式(左右翻轉、平移),分別還原這四張機密影像;其中,所需空間比所隱藏的機密影像總量還小。有條件的(2,3)型視覺加密,將四張機密影像編碼後產生三張投影片。其中,只有取得最重要的投影片時才可與其餘二張投影片中的任一張相互重疊還原機密影像;另外,我們將其作延伸,可變為(2,k)型。最後,在對折型中,也是產生二張投影片。要將其中一張投影片上下對折後與另外一張投影片重疊,才可還原資訊。
This master’s thesis proposes three schemes on visual cryptography of multiple secret images sharing: (2,2) model, constrained (2,3) model, and folding model. In the (2,2) model, we present two different schemes : 1) Multiple types of stacking: this method encodes twelve secret images into two transparencies. We can get twelve secret images respectively by eight different stacking types. 2) High hiding capacity: for four given secret images, two transparencies are produced. Stacking in different manners, we can decode four secret images respectively. Total space needed is smaller than the bits of total secret images. In constrained (2,3) model, the method encodes four secret images into three transparencies. We only can decode secret images by stacking the most important transparency with either one of the two other transparencies. This method can be extended into (2,k) model. Finally, in the folding model, we fold down a transparency, and then stack with the other to decode secret images.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT079957539
http://hdl.handle.net/11536/74347
Appears in Collections:Thesis