大部分浮水印的技術，要求取回嵌入的資訊時，不需要使用到原始的資料，這就是所謂的blind watermarking技術。本文提出一種簡單且有效的blind watermarking技術，在取出浮水印過程中不需原始影像與位置資訊。
本文中採用H.264的轉換演算法，即整數離散餘弦轉換(IntDCT)，以4x4的區塊大小進行轉換，可減低影像區塊效應的程度，藉以提高影像的品質。
在本文提出的方法中，分別將強健性浮水印與錯誤偵測資訊，嵌入至DC與低頻AC中；強健性浮水印的部分，為了保持好的影像品質，在嵌入過程中，微量的修改DC值；錯誤偵測資訊嵌入的方法則是利用相鄰區塊之DC係數，估測並修改中心區塊之低頻AC係數。
本論文將估測的方式應用到頻域中，能確保嵌入的浮水印在空間域中是不可見的。在取出的過程中，可以透過原先嵌入的錯誤偵測的資訊，指出影像被惡意竄改或資訊流失的部分；在取出浮水印時，即使影像遭受到破壞，利用錯誤偵測資訊，可在影像中拼湊出正確的浮水印。實驗結果顯示，本文所提出的浮水印系統具有高度的強健性，它能抵抗壓縮、切割、雜訊、模糊、亮度對比等各類訊號處理的攻擊。

For most watermark applications, it is often desired to retrieve the embedded information without access to the host data, which is a so-called blind watermarking technique. Therefore, a simple and effective blind watermarking scheme is expected. In this paper, we present a kind of blind watermarking technique.
The H.264 translate algorithm is used in this thesis. The Integer Decrease Cosine Transform (IntDCT), which uses 4x4 block to translate, is able to reduce the block effect.
In the proposed method, we embedded the robust watermark and information of error detection into DC component and low frequency AC coefficients of DCT, respectively. For the robust watermark, we modify the DC coefficients slightly on embedding process. For the information of error detection embedding, we use the DC coefficients of neighbor blocks to estimate and modify AC coefficients of central block.
We use the estimation in the frequency domain, and with that we can ensure that the embedded watermark is invisible in the frequency domain. In the extraction, we use the embedded information of error detection to find the missing or altered data of image. Even if the image is attacked, we still can use the information of error detection to find the correct watermark in extraction. Experimental results demonstrate the watermarks embedded with the proposed algorithm are robust to various attacks.

第一章 序論 1
1.1 簡介 1
1.2 數位影像驗證技術 3
1.3 研究動機 7
1.4 論文架構 8
第二章 影像浮水印技術簡介 9
2.1 數位浮水印技術基本原理 9
2.2 各種影像攻擊方法介紹 12
2.3 數位浮水印打亂技巧 14
2.3.1 Torus Automorphism 15
2.3.2 Linear Feedback Shift Register(LFSR) 15
2.4 離散餘弦轉換(DCT) 17
2.5 整數離散餘弦轉換(IntDCT) 19
第三章 相關數位浮水印技術 22
3.1 Choi and Aizawa 提出的方法 24
3.2 Yulin Wang 提出的方法 28
第四章 提出之數位影像驗證技術 32
4.1 浮水印結構 32
4.1.1 錯誤修正(Error-Correcting) 33
4.2 浮水印嵌入系統 35
4.2.1 浮水印嵌入演算法 36
4.2.2 DCT域低頻AC係數估測方法 38
4.2.2.1 IntDCT域低頻AC係數估測方法 40
4.2.3 錯誤偵測資訊嵌入系統 42
4.3 浮水印萃取系統 43
第五章 系統評估 46
5.1 影像品質評估 46
5.1.1 Peak Signal to Noise Ratio (PSNR) 46
5.1.2 Correct Decoding Rate 47
5.2 實驗結果 48
5.2.1 各式攻擊的錯誤偵測 53
5.2.2 各式攻擊的強健性 56
5.2.3 JPEG壓縮 63
第六章 結論與未來展望 64
6.1 結論 64
6.2 未來展望 65
參考文獻 66

圖目錄

圖2.1 典型浮水印嵌入法流程圖 10
圖2.2 私密浮水印萃取法流程圖 10
圖2.3 公開浮水印萃取法流程圖 11
圖2.4 利用亂數種子將浮水印打亂 14
圖2.5 利用亂數種子還原經亂數打亂的浮水印資料 14
圖2.6 LFSR 16
圖2.7 二維影像頻率分佈圖 18
圖2.8 (a)二維影像邊緣分佈圖(b)二維影像能量分佈圖 18
圖3.1 浮水印嵌入程序 25
圖3.2 V表示圖 26
圖3.3 測試影像Splash與Baboon 27
圖3.4 估測區塊(a)3*3,(b)5*5,(c)7*7,(d)9*9 29
圖3.5 使用鄰近區塊DC值估測中心區塊AC值示意 30
圖3.6 測試影像Lenna與Baboon 31
圖4.1 浮水印嵌入演算法流程圖 35
圖4.2 量化後F(0,0) 的範圍 37
圖4.3 量化後F(0,0) 所能承受的攻擊範圍 37
圖4.4 中心區塊(灰色)與相鄰區塊 39
圖4.5 (a)原始影像，(b)經估測後產生的新影像 39
圖4.6測試影像在不同a 下的PSNR結果 41
圖4.7 (a)原始影像，(b)經估測後產生的新影像 41
圖4.8 浮水印萃取流程圖 43
圖5.1 原始影像 48
圖5.2 浮水印(TKU)  49
圖5.3 包含浮水印影像 53
圖5.4 惡意竄改攻擊(a)偽裝(b)修改局部內容 54
圖5.5 惡意竄改攻擊偵測結果 54
圖5.6 經過壓縮(60%)與加入物件(直昇機)兩種攻擊 55
圖5.7 不同p 值偵測的結果 55
圖5.8 p=5  時，不同壓縮品質下偵測的結果 56
圖5.9 不同攻擊後的Lena影像-(1) 57
圖5.10 不同攻擊後的Lena影像-(2) 58
圖5.11 萃取浮水印採用錯誤修正碼採用多數決定法的結果 59
圖5.12 使用多數決定法萃取出之浮水印之 值 60
圖5.13 萃取浮水印採用錯誤修正碼採用選擇性決定法的結果 61
圖5.14 使用選擇性決定法萃取出之浮水印之 值 61
圖5.15 錯誤偵測參考影像，p=5 偵測結果-(1) 62
圖5.16 錯誤偵測參考影像，p=5 偵測結果-(2) 62
圖5.17 不同壓縮品質下取出之浮水印 值 63

表目錄

表3.1 影像品質與浮水印容量 27
表3.2 影像品質與浮水印容量 31
表5.1 4＊4 IntDCT方法之PSNR值 50
表5.2 浮水印容量 52
表5.3 包含浮水印影像之PSNR值 52

[1] N. Ahmidi, R. Safabakhsh, ”A novel DCT-based approach for secure color image watermarking,” ITCC 2004, vol. 2, pp. 709-713.
[2] R. B. Wolfgang, C. I. Podilchuk, E. J. Delp, ”Perceptual watermarks for digital images and video,” Proceedings of the IEEE, Vol. 87, No. 7, July. 1999, pp. 1108-1126.
[3] H. Kill, J. Onishi, S. Ozawa, ”The digital watermarking method by using both patchwork and DCT,” IEEE Multimedia Computing and System, Jun. 1999, pp. 895-899.
[4] J. Zhagn, Anthony T. S. Ho, ”An efficient digital image-in-image watermarking algorithm using the integer discrete cosine transform (IntDCT),” ICICS-PCM 2003, vol. 2, Singapore, 15-18 Dec. 2003, pp. 1163-1167.
[5] I. J. Cox, J. Kilian, T. Leighton, and T. Shamoon, ”Secure spread spectrum watermarking for multimedia,” IEEE Trans. Image Processing, vol. 6, Dec. 1997, pp. 1673-1687.
[6] C.S. Lu, S. K. Huang, C. J. Sze, and H. Y. M. Liao, ”Cocktail watermarking for digital image protection,” IEEE Trans. Multimedia, vol. 2, Dec. 2000, pp. 209-224.
[7] C. T. Hsu and J. L. Wu, ”Multiresolution watermarking for digital image, ” IEEE Trans. Circuits Syst. Ⅱ, vol. 45, Aug. 1998, pp. 206-216.
[8] C. T. Hsu and J. L. Wu, ”Hidden digital watermarks in images,” IEEE Trans. Image Processing, vol. 8, Jan. 1999, pp. 55-68.
[9] J. Huang, Y. Q. Shi, and Y. Shi, ”Embedding image watermarks in DC components, ” IEEE Trans. Circuits System Video Technology , vol. 10, Sept. 2000, pp. 974-979.
[10] P. H. W. Wong, O. C. Au, and J. W. C. Wong, ”Image watermarking using spread spectrum technique in log-2-spatio domain,” in Proc. IEEE Int. Symp. Circuits and Systems, vol. 1, May 2000, pp. 224-227.
[11] C. S. Lu and H. Y. M. Liao, ”Multipurpose watermarking for image authentication and protection,” IEEE Trans. Image Processing, vol. 10, Oct. 2001, pp. 1579-1592.
[12] J. R. Hernandez, F. P. Gonzalez, J. M. Rodriguez, and G. Nieto, ” Performance analysis of a 2-D-multipulse amplitude modulation scheme for data hiding and watermarking of still images,” IEEE J. Select. Areas Commun., vol. 16, May 1998, pp. 510-524.
[13] J. R. Hernandez, M. Amado, and F. Perez-Gonzalez, ”DCT-domain watermarking techniques for still image: detector performance analysis and a new structure,” IEEE Trans. Image Processing, vol. 9, Jan.2000, pp. 55-68.
[14] G. C. Langelaar and R. Lagendijk, ”Optimal differential energy watermarking of DCT encoded images and video,” IEEE Trans. Image Processing, vol. 10, Jan. 2001, pp. 148-158.
[15] P. W. Wong and N. Memon, ”Secret and public key image watermarking schemes for image authentication and ownership verification,” IEEE Trans. Image Processing, vol. 10, Oct. 2001, pp. 1593-1601.
[16] Y. J. Zhagn, T. Chen, and J. Li, ”Embedding watermarks into both DC and AC components of DCT,” in Proc. SPIE Security and Watermarking of Multimedia Contents Ⅲ, Jan. 2001, pp. 424-435.
[17] S. Stankovic, I. Djurovic, and I. Pitas, ”Watermarking in the space/spatial-frequency domain using two-dimensional Radon-Wigner distribution,” IEEE Trans. Image Processing, vol. 10, Apr. 2001, pp. 650-658.
[18] W.N. Lie, G. S. Lin, C. L. Wo, and T. C. Wang, ”Robust image watermarking on the DCT domain,” in Proc. IEEE Int. Symp. Circuits and Systems, vol. 1, May 2000, pp.228-231.
[19] Y. Choi, I. Aizawa, ”Digital watermarking using inter-block correlation,” in Proc. International Conf. on Image Processing, vol. 2, October 1999, pp.216-220.
[20] Y. Wang, A. Pearmain, ”Blind image date hiding based on self reference,” Pattern Recognition Letters 25, Aug. 2004, pp. 1681-1689.
[21] C. A. Gonzales, L. Allman, T. Mccarthy, P. Wendt, ”DCT coding for motion video storage using adaptive arithmetic coding,” Signal Processing: Image Comm. 2, 1990, pp.145-154.