基於分享影像尺寸不變的灰階與彩色機密影像分享技術之研究

A Study of Share Size Invariant Secret Sharing Techniques for Gray-scale and Color Images

游瑞文(Rui-Wen Yu)

指導教授 ： 吳憲珠

國立臺中科技大學/資訊學群/資訊科技與應用研究所/碩士(2009年)

摘要

視覺機密分享技術是一個提供多人分享機密影像的安全機制；藉由特殊的編碼方式可將機密影像之資訊藏入於多張影像內（稱之為分享影像），日後欲解開機密資訊時，只需藉由分享影像的疊合即可顯示所藏入之機密資訊，而不須透過複雜之運算來解密。近年來，此技術已廣泛延伸至灰階機密影像及彩色機密影像的應用。然而，大多數的視覺機密分享技術所產生的分享影像皆會造成影像尺寸擴張的現象，進而造成網路頻寬與儲存空間的負擔。為了改善此問題，本論文主要針對視覺機密分享於分享影像尺寸不變量之研究進行探討。
首先，本論文於第三章中，針對灰階機密影像提出一個產生尺寸不變之分享影像並疊合後具有高影像品質之機密影像分享機制。此方法主要是利用具特徵樣式直方圖和具多色階編碼矩陣來提升還原後的機密影像品質。此外，利用“區塊對應區塊”的編碼方式來維持分享影像尺寸的不變性。實驗結果顯示，本技術能夠藉由提供多色階的方式使得還原後的機密影像擁有較佳的視覺品質，並且改善影像資訊流失的問題。
在第四章中，本論文提出一個有意義的彩色機密影像分享的機制。此技術主要利用半色調技術與色彩最佳化的方式來降低彩色機密影像的資訊量，並將機率視覺機密分享機制擁有像素不擴張的特性延伸至彩色機密分享機制，使得分享影像能夠維持尺寸的不變性。此外，為了減少分享影像於傳輸過程中受到駭客關注的可能性，本技術將機密影像藏入於自然影像中達到偽裝保護性。實驗結果顯示，彩色機密影像能夠藉由疊合足夠數量的分享影像予以還原。此外，本技術除了改善先前彩色機密影像分享機制的共通問題：像素擴張，其所提出之彩色機密影像分享架構亦能夠適用於二種色彩模式。
最後，於第五章中提出一個具有驗證功能之彩色機密影像分享機制。首先，為了使彩色機密影像易於編碼處理，本技術藉由前處理來降低彩色影像的資訊量。接著，亦同樣地延伸機率視覺機密分享機制至本技術，以保持分享影像尺寸的不變性。此外，對機密影像編碼的同時，驗證資訊亦被編碼入於分享影像而不造成其影像尺寸之擴張。實驗結果顯示，每位機密分享者皆可與任一機密分享者透過位移式疊合其雙方所持有之分享影像，並從疊合之影像區域來顯現驗證資訊，進而辨別分享影像的真偽。此外，彩色機密影像不需任何複雜的計算僅藉由疊合分享影像的方式便能清楚地還原其資訊內容。

關鍵字

視覺機密分享 ； 視覺密碼學 ； 分享影像尺寸不變

並列摘要

Visual secret sharing (VSS) is an efficient way to share a secret image among many participants. In this scheme, the secret image is encrypted into many images which are also called shares. The secret image is revealed by overlapping share images without any complex computation. In recent years, this scheme has been employed to deal with gray-scale and color images for enhancing applications of the sharing way. However, most schemes result in the generated shares with large size expansion to cause the burden of bandwidth and storage space. In order to improve this drawback, this thesis focuses the study on visual secret sharing with share size invariant.
First, a scheme based on gray-scale image to provide a high quality VSS scheme with share size invariant is presented in Chapter 3. In this scheme, the pattern histogram and MLB matrices are employed to enhance the quality of the reconstructed secret image. The encrypting way “block by block” is employed to maintain the share size invariant. Experimental results show that the proposed scheme can provide a better quality for the reconstructed secret image by improving the problem of information lost and providing more than two color levels for shares.
In Chapter 4, a (k, n)-threshold color visual cryptography (CVC) with meaningful shares is proposed. In this scheme, halftone technique and optimizing color is employed to reduce the color information of a secret color image. Then, the secret is encrypted by the concept of probabilistic VSS for maintaining the share size invariant. In the meantime, the shares are embedded in nature images for reducing the attention of hackers. Experimental results show that the proposed scheme improves the shortcoming of previous color visual cryptography studies in pixel expansion. The secret image is revealed by overlapping sufficient meaningful shares. Moreover, this proposed scheme can be applied to additive and subtractive color models.
Finally, a CVC scheme with authentication information is proposed in Chapter 5. In this scheme, the secret color image is processed by pre-processing for reducing the information size. Then, the secret image is encrypted by a probability CVC scheme. At the same time, the authentication information is embedded in each share. Experimental results show that each participant can identify each share owned by other participants. Moreover, the secret color image can be clearly revealed without any computation, and the size of each share is invariable.

並列關鍵字

Visual secret sharing ； visual cryptography ； share size invariant

參考文獻

[1] G. Ateniese, C. Blundo, A. De Santis, and D. R. Stinson, “Visual Cryptography for General Access Structures,” Information and Computation, Vol. 129, 1996, pp. 86-106.

[3] M. Benaissa and W. M. Lim, “Design of Flexible GF(2m) Elliptic Curve Cryptography Processors,” IEEE Transactions on Very Large Scale Integration Systems, Vol. 14, 2006, pp. 659-662.

[4] C. Blundo and A. De Santis, “Visual Cryptography Schemes with Perfect Reconstruction of Black Pixels,” Computers & Graphics, Vol. 22, No. 4, 1998, pp. 449-455.

[5] C. Blundo, P. D’Arco, A.D. Santis, and D.R. Stinson, “On the Contrast in Visual Cryptography Schemes,” Journal of Cryptology, Vol.12, 1999, pp.261-289.

[6] C. Blundo, P. D’arco, A.D. Santis, and D.R. Stinson, “Contrast Optimal Threshold Visual Cryptography,” SIAM Journal on Discrete Mathematics, Vol.16, 2003, pp.224-261