摘要: | 在現行漸進式視覺密碼的研究假設中,是讓每一張分享影像都擁有相同的權限等級,因此在這個限制下,管理人員無法根據參與者的重要性,針對機密影像的還原而給予差異化的能力。為了改善這個限制,本研究提出漸進式且具有不同權限等級的不擴展視覺密碼分享方法,稱之為 (n, n)-priority and progressive visual sharing model ((n, n)-PPVSM)。在這個分享模型中,假設共有 n 位機密資訊分享的參與者,並且每一位參與者的權限都不相同,因此每一張分享影像都被賦予一個不同的還原機密影像的權限等級 (k)。
在本研究的設計中,第 k 個權限等級的分享影像 (WSk) 上,每一個像素點被分配到黑點的機率是 2k / (n2 + n),並且如果權限等級 k 大於權限等級 l,則 WSk 還原機密影像的能力將優於 WSl。當重疊所有的分享影像後,機密影像的白點部分被疊合出黑點的機率為2 / (n + 1),而黑點部分則是會被疊合出全黑,在疊合影像上會產生 (n − 1) / (n + 1) 的黑白色差,所以可以清晰地顯示機密影像的內容。
相較於其它學者的研究,(n, n)-PPVSM 的分享機制具備下列幾項優點:(1) 每一張分享影像具備不同的能力來還原機密影像。(2) 疊合影像的色差對比優於其它視覺密碼學的研究。(3) 權限機制更適合應用於灰階和彩色影像的漸進式分享環境中,可以清晰地辨識出機密內容。(4) 本研究的分享影像大小與機密影像相同。 In the recent studies of visual cryptography (VC) or progressive VC, every share created by above two schemas has same capabilities to recover the secret image. As a result, managers are unable to give proper advantage to participants according to their importance. Consequently, providing each share a suitable recovering capability will be an effective way to solve this problem. In this paper, we propose a novel secret sharing method namely (n, n)-priority and progressive visual sharing model ((n, n)-PPVSM), every participant will be assigned a distinct weighting value, assuming that each one's significance is different from each other.Particularly, every pixel has the probability of 2k/(n^2+n) to be distributed as black in the k-th weighting share (WS(superscript k)), and this share has a better ability than the l-th weighting share (WS(superscript l)) to reveal the secret image if k≧l. When superimposing all shares, the upper bound of the white part in the secret image to be appeared as a black point will be 2/(n+1), and the black part is fully black, the restored image's contrast is equal to (n-1)/(n+1), and the contents will be clearly identified.Comparing our work with others, (n, n)-PPVSM has the following advantages: (1) Every share has different abilities to reveal secret information. (2) The contrast of the restored image is better than traditional VC. (3) This mechanism is more suitable for gray-level and color images in progressive sharing, and the cryptographers can clearly identify confidential content visually. (4) The size of shares is the same as the one of the secret images. |