應用新型視覺密碼的強健型浮水印於數位使用權認證研究

A Study on Digital User Authentication Using New Robust Watermarking Based on Visual Cryptography

蔡東霖(Tung-Lin Tsai)

指導教授 ： 駱榮欽

國立臺北科技大學/電資學院/電腦與通訊研究所/碩士(2007年)

摘要

在本論文中，我們提出新的基於視覺密碼學(Visual Cryptography)的浮水印保護技術。在過去有很多的浮水印研究被提出，這些技術不外乎是空間域或頻率域的處理。空間域的技術運算量低易執行，但強健性不足;而頻率域的技術運算量高，大多具有足夠的強健性。在本論文提出的方法，不但有較低的運算量而且更有高強健性的優點。
視覺密碼(Visual Cryptography) [1] 在1994年由Noar與Shamir所提出的，是對影像加密的一個方法。利用視覺密碼來對機密影像加密後，可以得到兩張分享影像，破密者無法經由單張分享影像的分析，得到機密影像的任何訊息。而解密的方法相當簡單，不需要複雜的電腦運算；只要將分享影像一(share1)疊放在另一張分享影像二(share2)之上，即可以看到機密影像。
我們將應用視覺密碼技術(VCS)將浮水印加密成兩張分享影像。其中一張藏入需保護的影像中作為認證的依據；而另一張則作為解密所需的密鑰。在實驗中，我們將黑白、灰階及全彩水印各藏入灰階及全彩影像中，並模擬在受各種攻擊的情況下，取出的水印的辨視率相當的高，可證明能強健的抵擋各種攻擊。

關鍵字

資訊隱藏 ； 浮水印 ； 視覺密碼

並列摘要

In this thesis, a new construction algorithm of watermarking based on visual cryptography is proposed. In the past years, there are many watermarking schemes addressed, these schemes most include spatial-domain and transform-domain. Spatial-domain is easy to access and low computing load but robust not enough. Transform-domain has good robust application, but the computing load also high the former.
Visual cryptography is a new cryptography idea for image encryption, which was addressed from M. Naor and A. Shamir in 1994 [1]. After utilizing the visual cryptography to encrypt the secret image, we can get two shares. The persons who decode are unable via the single share image to get any information of the secret image. And the method to decipher is quite simple, do not need complicated computer operation; So long as one share (share1) will be stacked in second share (share2) another one, namely can see secret image.
We will use visual cryptography scheme (VCS) to encrypt watermark into two shares. One is embedded into protected image as authentication; another one is regarded as and deciphered the necessary key. Among experiments, we try to use two-tone, gray-level and color watermarks to embed into gray-level and color images, and simulation is in case of receiving various kinds of attacks. The watermark is retrieved then distinguished high by the human visual system; we can prove one thing that endures various kinds of attacks strongly.

並列關鍵字

Information hiding ； Watermarking ； Visual cryptography

參考文獻

[2] M. Al-khassawneeh and S. Aviyente, "Robust watermarking on the joint spatial-spectral domain," IEEE Signal Processing Education Workshop, pp. 297 - 301, 2004.

[3] A. Miyazaki and A. Okamoto, "Analysis of watermarking systems in the frequency domain and its application to design of robust watermarking systems," IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 3, pp. 1969 - 1972, 2001.

[4] I. J. Cox, J. Kilian, F. T. Leighton, and T. Shamoon, "Secure spread spectrum watermarking for multimedia," Image Processing, IEEE Transactions, vol. 6, pp. 1673 - 1687 1997.

[5] A. B. Piva, F., Cappellini, V. and Barni, M., "Exploiting the cross-correlation of RGB-channels for robust watermarking of color images " International Conference on Image Processing, vol. 1, pp. 306 -310, 1999.

[6] X.-M. Niu, Z.-M. Lu, and S.-H. Sun, "Digital watermarking of still images with gray-level digital watermarks," Consumer Electronics, IEEE Transactions, vol. 46, pp. 137 - 145 2000.