由於網際網路技術的快速發展使得數位多媒體資訊的傳輸變得更加容易與快速，也因此如何保護資訊傳輸的安全確保智慧財產權的完整，目前已成為熱門研究的課題。而數位浮水印技術提供了解決這類問題的方法。主要是在數位多媒體藏入不可視的資料，當有人竊取時就可還原浮水印，以利驗證。但是藏入的數位浮水印必須具備對原始影像要低破壞、安全性，和強韌性的特性。

本論文中，我們提出數位浮水印搭配均勻與非均勻量化做結合的技巧，也就是被量化點必須配合應植入的數位浮水印而量化到該重建值的位子上，我們稱做是浮水印量化。在數位浮水印植入方面，利用展頻觀念，將ㄧ個數位浮水印位元的資訊放在數個像素上，如此可抵抗雜訊攻擊，以保護數位浮水印的原始資料。且在浮水印量化的部份，有使用以1為量化移動單位的均勻量化，還有在非均勻量化我們加入Lloyd-Max量化器來搭配影像的新方法，運用其差值來當量化的移動單位，完成植入的動作。本論文推導出因雜訊而產生的數位浮水印錯誤率，以及與數位浮水印有關的總影像失真量，讓我們在未真正植入前，即可預知當數位浮水印放入後，對影像品質和浮水印的評估。由實驗結果可證明我們估測的效能是正確的。本論文所提出的方法也確實能達到著作權保護、秘密通訊的目的。

Due to rapid evolution of Internet technology, the transmission of digital multimedia becomes easier and fast. How to protect the security of information transmission and make sure the completeness of copyright is becoming a popular research topic in recent years. Digital Watermarking provides a good way to solve this kind of problem. It achieves its goal by hiding the copyright information invisibly in multimedia data. If someone stole the data, we can extract the watermark and assert that the copyright belong to us. Digital watermarking should induce low impairment to the original image. It should also provide security and be robust to tampering.

In this thesis, we propose a watermarking scheme that is merged into image’s quantization, which can be uniform or non-uniform. In the case of non-uniform quantization, we assume that the Lloyd-Max quantizer is adopted. We use the spectrum spreading to embed. The effect of embedding one watermark bit is spread over a sequence of pixels selected by a PN code. The values of those pixels are either unchanged or slighted modified (to their neighboring quantization values) so that they sum up (weighted by a code sequence of alternating +1 and -1) to indicate whether the watermark bit is -1 or 1. In the embedding process, a few parameters can be chosen to control the tradeoff between the robustness of watermarks. We are able to derive explicit formulae, which involve those embedding parameters, for the distortion to the original image caused by the watermark embedding and the WBER (watermark bit error rate) when the watermarked image is tampered by Gaussian noise. Experiments show that the theoretical results predicted by those formulae are in good agreement with the experimental results. The experimental results reveal the proposed method could successfully ensure for the copyright protection and secret information communication.

[1] Lv Jiu-ming. , “Digital Watermark technique aiid its attack metliod
based 011 military communication [C] ,” Thesis collection of anti-
jamming in militaq comiunication 2003, Hefei, 2003.11: 128-134

[2] S.-J. Lee S.-H. Jung, “A survey of watermarking techniques applied to
multi-Media, ” Proc. Of IEEE Inter. Symposium on Industrial Electronics,
Vol. 1, pp.272 – 277, June 2001.

[3] Pei-Min Chen,“A Visible Watermarking Mechanism Using a Statistic
Approach” Proc. IEEE, Vol. 2, pp.910-913, Aug. 2000

[4] Y. Hu, S. Kwong, “An image fusion based visible watermarking algorithm,”
ISCAS ’03, Proc. Of the 2003 Inter. Symposium on Circuits and System,
Vol. 3, pp. III – 794 – III – 797, May 2003.

[5] Shih-Hao Wang, Yuan-Pei Lin, “Blind Watermarking Using Wavelet Tree
Quantization”; Proc.IEEE,Vol.1,pp. 589 – 592,Aug. 2002

[6] J.Herrera-Joancomarti,J.Minguillon,and D.Megias,“ A Family of Image
Watermarking Schemes Based on Lossy Compression” Proc. IEEE. International
conference on, pp. 559 - 563 , April 2003.

[7] Yazhou Lin,Wen Gao,Hongxun Yao,and Shaohui Lin,“A Texture-Based Tamper
Detection Scheme by Fragile Watermark” Proc.IEEE,Vol.2,pp.177-180,May
2004.

[8] Alghnniemy M. and Tedik A., I‘ Progressive Quantized Projection
Watermarking Scheme”, Proc. the 7th ACM MM99 Infl. Multimedia Con$, pp.
295-298,Orlando, Nov. 1999.

[9] D. Karakos and A. Papamarcou, “A relationship between quantization and
distribution rates of digitally watermarked data,” Univ. Maryland, Inst.
Syst. Res. Tech. Rep TR 2000-51.
[10] Bingxin Li and Hannu Tenhunen, “Sigma delta modulators using semi-
uniform quantizers,” Proc. IEEE, Vol. pp.456 - 459, May 2001.

[11] Nin.9 Lau and K.P. Subbalak.slima, “NON-UNIFORM QUANTIZER DESIGN
FOR IMAGE DATA HIDING,” 2004 International Conference on Image
Processing, S t w"v Inst i tut e of Technology, Hohoken, NJ 07030, USA

[12] A. Ghassemi, T.A. Gulliver, “Turbo decoding with nonuniform
quantization,” Proc. IEEE, Vo3. pp. 2273 – 2277, Sept. 2004.

[13] J. Max,“Quantizing for minimum distortion,” IRE Trans. Inform. Theory,
IT-6, pp. 7-12, Jan. 1960.

[14] S. P. Lloyd, “Least squares quantization in PCM,” IEEE Trans. Inform.
Theory, Vol. 28, pp. 127-135, Mar. 1982.

[15] A.T.S. Ho, F. Shu, “A robust spread-spectrum watermarking method using
two- level quantization, ” Proc. IEEE, Vo2. pp.725 - 728, Oct. 2004.

[16] Cheng – Chung Sun, Secure Spread Spectrum Watermarking base on Trellis
Coding, Master’s thesis of NTUST, Taipei, ROC, 2003.

[17] Guixing Wu, En-Hui Yang, “Joint watermarking and compression using
scalar quantization for maximizing robustness in the presence of additive
Gaussian attacks. ” Digital Object Identifier 10.1109/TSP.2004, pp.834 -
844, Feb. 2005.

[18] Chun-Shien Lu, Hong-Yuan Mark Liao and Chwen-Jye Sze, “Combined
Watermarking for Image Authentication and Protection “, IEEE Int. conf.
on Multimedia and Expo, Vol. 3, pp.1415-1418, 2000.

[19] Ping Wah Wong and Nasir Memon, “Secret and Public Key Image Watermarking
Schemes for Image Authentication and Ownership Verification”, IEEE Trans.
on Image Processing, Vol.10, pp. 1593-1601, 2001.

[20] K. Ramakrishnan, A. Balasubramanian, S. Mishra, R. Sridhar , “Wireless
Security Protocol using a Low Cost Pseudo Random Number Generator,” Proc.
IEEE, pp. 17-20, Oct. 2005.

[21] K.H. Tsoi, K.H. Leung and P.H.W. Leong, “Compact FPGA-based True and
Pseudo Random Number Generators,” Proc. Of the 11th Annual IEEE
Symposium on Field-Programmable Custom Computing Machines (FCCM’03),
2003.