以隨機網格為基礎的視覺機密分享

機構典藏 > College of Business and Management > Graduate Institute & Department of Information Management > Research Paper > Item 987654321/102902

Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/102902

Title:	以隨機網格為基礎的視覺機密分享
Other Titles:	Random Grid Based Visual Secret Sharing
Authors:	侯永昌
Contributors:	淡江大學資訊管理學系
Keywords:	隨機網格;有意義偽裝分享影像;視覺機密分享;Random Grid;Friendly Shares;Visual Secret Sharing
Date:	2012-08
Issue Date:	2015-05-11 15:20:32 (UTC+8)
Abstract:	隨機網格視覺機密分享 (random grid visual secret sharing，RGVSS)是由 Kafri 與 Keren (1987) 所提出的，所謂隨機網格的意思就是每一個網格內容必須符合隨機變數的要求，也就是服從統計學上獨立且分配一致 (independent and identically distributed， IID) 的要求。隨機網格的優點是不需要建置矩陣，並且所產生的分享影像皆與機密影像一樣大，因此這是一項成本低且簡單的加密技術。 Shyu (2007, 2009) 擴充了 Kafri 與 Keren 的作法，提出了三種不同透光度 (視覺效果) 的隨機網格模型。但是隨著參與者和被重疊分享投影片數目的增加，機密影像的白點部份，在疊合影像上仍是出現白點的機率將會下降，使得三種分享方式在疊合影像上的色差對比值分別下降到 1/2n-1、1/(2n+1) 和 1/2n，因此造成了還原品質不佳的問題。 Chen and Tsao (2011) 應用 Shyu (2007) 所提出的三個模型，提出一個有意義且非擴展的視覺機密分享模型，稱之為 friendly RGVSS (FRGVSS)。但是他的做法至少有下列三項缺點：在分享影像與疊合影像的色差都值比較差、分享影像只能是同一個影像，且在極端情況下，疊合影像上將無法看出機密影像的內容等問題。本研究將針對 Chen and Tsao (2011) 的缺點提出一個以隨機網格為基礎的視覺機密分享方法，我們的方法所產生的分享影像將會是不同的掩護影像，我們將使用全部的像素位置來產生分享影像，並將機密影像的資訊隱藏在分享影像中，使得分享影像與疊合影像的色差都達到理論上的極大值。 Kafri and Keren (1987) first introduced the use of random grids in their encryption method. The content assignment of every random grid should satisfy the requirements of random variables, which is to obey the statistical criteria, to be independently and identically distributed (IID). Thus, the probability of black and white pixels appearing in each grid would be the same, both 1/2. The size of each share will be the same size as the secret image. It is a simple and useful technique to share secret image. Shyu (2007, 2009) used this idea to generate non-expanded shares. The first step is to set a random grid r1 for the first share; then, based on the probability theorem, they designed three algorithms to encrypt the other shares. However, as the number of overlapped shares increases, the probability of having white pixels in the white blocks of the secret image will become lower and lower. This also means that the light contrast (LC) will gradually decline. Therefore, in his VCRG-n sharing model, the LC values will be reduced to 1/2n-1, 1/(2n+1) and 1/2n, respectively. Hence, the stacked image will not have sufficient contrast, which causes the recovered image to be of poor quality. the light contrast values of the three algorithms Chen and Tsao (2011) adopted the models proposed by Shyu (2007) and presented a user-friendly random-grid-based visual secret sharing scheme, friendly RGVSS (FRGVSS). But the have 3 drawbacks: the contrasts of the shares and the stacked image are poor, the contours on each share are the same, and in some worst cases, no secret image will be shown on the stacked image. Our proposed scheme can overcome Chen and Tsao’s (2011) drawbacks and may have the following advantages: (1) each share can use different image as its cover image; (2) the contrasts of the shares and the stacked image will reach their theoretical optimum values, hence we can get better visual quality; (3) the share size is the same as that of the secret image.
Appears in Collections:	[Graduate Institute & Department of Information Management] Research Paper

Files in This Item:

There are no files associated with this item.

Loading...