臺灣博碩士論文加值系統

本論文提出將浮水印資訊嵌入於DCT域中低頻AC係數之浮水印法，因為不可視性為浮水印技術中相當重要的特性，因此在嵌入浮水印的過程中保留DC係數值，避免導致影像品質低落。在本論文所提出的四種嵌入方法中，利用鄰近區塊的DC係數，預測中心區塊的AC係數，藉由調整AC係數值嵌入浮水印資訊。並利用LMS演算法訓練出濾波器係數，配合鄰近區塊的DC係數，能精準的預測原始影像的AC係數，故能有效的萃取出浮水印資訊。
　　本論文所提出的四種方法分別為：WEPAC (Watermark Embedding in the Predicted AC Coefficient)，顧名思義就是將浮水印嵌入於預測後的AC係數中；WEOAC (Watermark Embedding in the Original AC Coefficient)，將浮水印直接嵌入於原始影像的AC係數中；VWPAC (Variance-based Watermarking in the Predicted AC Coefficient)，就是以變異量分類為基礎，將浮水印嵌入於預測後的AC係數中；VWOAC (Variance-based Watermarking in the Original AC Coefficient)，同樣先經分類後，再將浮水印直接嵌入於原始影像的AC係數中。這四種方法各有不同的優缺點，可分別應用在不同的場合中。本論文也將利用各種不同程度的攻擊來測試其強健性。

This study presents a robust and semi-blind watermarking by embedding information into low frequency AC coefficients of Discrete Cosine Transform (DCT). Since the imperceptibility is the most significant issue in watermarking, the DC value is maintained unchanged. The proposed methods utilize the DC values of the neighboring blocks to predict the AC coefficients of the center block. The low frequency AC coefficients are modified to carry watermark information. The Least Mean Squares (LMS) is employed to yield the intermediate filters, cooperating with the neighboring DC coefficients to precisely predict the original AC coefficients.
Four watermarking methods, namely Watermark Embedding in the Predicted AC Coefficient (WEPAC), Watermark Embedding in the Original AC coefficient (WEOAC), Variance-based Watermarking in the Predicted AC Coefficient (VWPAC) and Variance-based Watermarking in the Original AC coefficient (VWOAC), are presented in this study. These methods are having their own advantages and disadvantages, and could be used in different applications. Moreover, many attacks are addressed to show the robustness of the proposed methods.

指導教授推薦書
口試委員會審定書
授權書
誌謝
中文摘要
英文摘要

目錄
第一章　緒論 1
1.1背景說明與動機 1
1.2研究目的 2
1.3論文架構 2
第二章　相關知識與文獻探討 4
2.1離散傅立葉轉換 (Discrete Fourier Transform, DFT) 4
2.2離散小波轉換(Discrete Wavelet Transform, DWT) 5
2.3離散餘弦轉換 (Discrete Cosine Transform, DCT) 6
2.4可視的浮水印技術 (Visible Watermarking) 7
2.5不可視的浮水印技術 (Invisible Watermarking) 8
2.5.1還原浮水印不需原始影像 (Blind Decode) 8
2.5.2還原浮水印不需原始影像，但需附加資訊 (Semi-blind Decode) 10
2.5.3 還原浮水印需利用原始資訊 (Non-blind Decode) 12
2.5.4 空間域浮水印 12
2.5.5 頻率域浮水印 13
2.5.5.1 離散餘弦轉換域(DCT Domain) 13
2.5.5.2 離散小波轉換域 (DWT Domain) 13
2.6浮水印的認證方式 14
2.6.1 影像訊號雜訊比 (Peak Signal to Noise Ratios, PSNR) 14
2.6.2 浮水印正確的解碼率 (Correct Decoding Rates, CDR) 14
第三章　嵌入於預測後及預測前之AC係數浮水印法 15
3.1 最小均方根演算法 (Least Mean Square, LMS) 20
3.2 變異量(variance)區塊分類法 23
3.3 WEPAC與WEOAC嵌入法 25
3.4 浮水印萃取法 26
第四章　基於變異量嵌入於預測後及預測前的AC係數浮水印法 28
4.1　變異量(variance)區塊分類法 28
4.2 VWPAC與VWOAC嵌入法 34
4.3　浮水印萃取法 37
第五章　實驗結果 39
5.1 影像的不可察覺性 40
5.1.1 WEPAC與WEOAC嵌入法 40
5.1.2 VWPAC與VWOAC嵌入法 40
5.1.3 所提出方法之討論 47
5.2 嵌入法之強健性－未受攻擊之CDR 48
5.2.1 WEPAC與WEOAC嵌入法 48
5.2.2 VWPAC與VWOAC嵌入法 50
5.2.3 所提出方法之討論 54
5.3 嵌入法之強健性－剪裁(Crop)攻擊 55
5.4 嵌入法之強健性－銳利化(Sharping)攻擊 57
5.5 嵌入法之強健性－竄改(Tamper)攻擊 59
5.6 嵌入法之強健性－旋轉攻擊 61
5.7 嵌入法之強健性－壓縮攻擊 64
5.7.1 JPEG壓縮攻擊 64
5.7.2 JPEG2000壓縮攻擊 67
5.8實驗結果之探討 70
第六章　實驗結果之比較 72
　　6.1 四種嵌入法之比較 72
6.2 四種嵌入法與其他文獻之比較 72
第七章　結論 75

圖目錄
第一章　緒論 1
第二章　相關知識與文獻探討 4
圖2-1二階小波轉換後七個頻帶示意圖 5
圖2-2二維DCT的基底函數 7
圖2-3 LVQ浮水印架構 9
圖2-4 Hsu與Wu所提出的基於DCT的浮水印架構 9
圖2-5 Huang等人提出改善的浮水印架構 10
圖2-6 Cox等所提出的浮水印架構 11
第三章　嵌入於預測後及預測前之AC係數浮水印法 15
圖3-1 PSNR與LMS濾波器大小之關係圖 16
圖3-2　LMS演算法所訓練出的濾波器 18
圖3-3 (0,1) AC係數的預測方法 19
圖3-4 PSNR與不同係數的關係圖(50張重建影像平均) 19
圖3-5 PSNR、變異量與每一個AC係數間的關係 (50張測試影像之平均結果) 20
圖3-6 四種不同分類分別預測不同個數之AC係數與PSNR、CDR間的關係(平均50張重建影像的結果) 23
圖3-7 WEPAC 與 WEOAC嵌入法 25
圖3-8 萃取浮水印架構 27
第四章　基於變異量嵌入於預測後及預測前的AC係數浮水印法 28
圖4-1 影像區塊依變異量的分佈圖(量化到相等面積) 29
圖4-2 影像區塊類別個數與重建影像品質間的關係 (平均50張測試影像之結果) 29
圖4-3 LMS演算法所訓練出的濾波器 31
圖4-4 PSNR與前9個AC係數間的關係 32
圖4-5五種不同類別分別預測不同個數之AC係數與PSNR間的關係(平均50張重建影像的結果) 34
圖4-6 VWPAC 與 VWOAC 嵌入法架構 36
圖4-7 萃取浮水印架構 38
第五章　實驗結果 39
圖5-1 8張測試影像 39
圖5-2 影像不可察覺性比較圖 42
圖5-3 影像不可察覺性比較圖 43
圖5-4 影像不可察覺性比較圖 45
圖5-5 影像不可察覺性比較圖 47
圖5-6 WEPAC與WEOAC嵌入法在不同狀況下之PSNR比較圖 48
圖5-7 VWPAC與VWOAC嵌入法在不同狀況下之PSNR比較圖 48
圖5-8 萃取浮水印之CDR比較圖 50
圖5-9 萃取浮水印之CDR比較圖 50
圖5-10 萃取浮水印之CDR比較圖 52
圖5-11 萃取浮水印之CDR比較圖 53
圖5-12WEPAC與WEOAC嵌入法在不同狀況下之CDR比較圖 54
圖5-13VWPAC與VWOAC嵌入法在不同狀況下之CDR比較圖 55
圖5-14 經剪裁後的嵌入浮水印影像 55
圖5-15 WEPAC與WEOAC嵌入法經剪裁攻擊的結果 56
圖5-16 VWPAC與VWOAC嵌入法經剪裁攻擊的結果 57
圖5-17經銳利化後的嵌入浮水印影像 58
圖5-18 WEPAC與WEOAC嵌入法經銳利化攻擊的結果 58
圖5-19 VWPAC與VWOAC嵌入法經銳利化攻擊的結果 59
圖5-20 竄改樣板與竄改結果 60
圖5-21 WEPAC與WEOAC經竄改攻擊後之結果 60
圖5-22 VWPAC與VWOAC經竄改攻擊後之結果 61
圖5-23 浮水印影像經旋轉後之結果 62
圖5-24 WEPAC與WEOAC嵌入法經旋轉攻擊後之結果 63
圖5-25 VWPAC與VWOAC嵌入法經旋轉攻擊後之結果 64
圖5-26 不同影像品質的壓縮結果 65
圖5-27 WEPAC與WEOAC嵌入法經JPEG壓縮後之結果 66
圖5-28 VWPAC與VWOAC嵌入法經JPEG壓縮後之結果 67
圖5-29 不同壓縮倍率的壓縮結果 68
圖5-30 WEPAC與WEOAC嵌入法經JPEG2000壓縮後結果 69
圖5-31 VWPAC與VWOAC嵌入法經JPEG2000壓縮後結果 70
圖5-32 從10000張不同的浮水印影像找出正確的解碼率 71
第六章　文獻比較 72
第七章　結論 75

表目錄
第一章　緒論 1
第二章　相關知識與文獻探討 4
第三章　嵌入於預測後及預測前之AC係數浮水印法 15
第四章　基於變異量嵌入於預測後及預測前的AC係數浮水印法 27
第五章　實驗結果 39
表5-1 嵌入64x64大小浮水印後之PSNR值 40
表5-2嵌入128x128大小浮水印後之PSNR值 41
表5-3嵌入128x128大小浮水印後之PSNR值 43
表5-4嵌入256x256大小浮水印後之PSNR值 44
表5-5 萃取64x64大小浮水印之CDR值 49
表5-6萃取128x128大小浮水印之CDR值 49
表5-7萃取128x128大小浮水印之CDR值 51
表5-8萃取256x256大小浮水印之CDR值 51
第六章　文獻比較 72
表6-1 四種嵌入法之比較 72
表6-2 與Choi及Wang等提出的方法比較之結果 73
表6-3 與Choi提出的方法比較壓縮攻擊之結果 74
表6-4 與其他嵌入法特性比較表 74
第七章　結論 75

[1] A. B. Watson, G. Y. Yang, J. A. Solomon, and J. Villasenor, “Visibility of wavelet quantization noise,” IEEE Transactions on Image Processing, vol. 6, no. 8, pp. 1164-1175, Aug. 1997.
[2] W. K. Pratt, Digital Image Processing, Wiley-Interscience, New York, 1978.
[3] P. M. Chen, “A visible watermarking mechanism using a statistic approach,” International Conference on Signal Processing Proceedings, vol. 2, pp. 910-913, Aug. 2000.
[4] X. Q. Li and X. Y. Xue, “A novel blind watermarking based on lattice 　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　vector quantization,” Canadian Conference on Electrical and Computer Engineering, vol. 3, pp. 1823-1826, May 2004.
[5] S. F. Sun, P. L. Qiu, Y. L. Wang and L. F. Yao, “Blind watermarking algorithm based on general Gaussian model,” International Conference on Signal Processing, vol. 3, pp. 2302-2305, Aug. 2004.
[6] C. H. Huang, J. L. Wu and D. Y. Chen, “A blind watermarking algorithm with semantic meaningful watermarks,” the Thirty-Fourth Asilomar Conference on Signals, Systems and Computers, vol. 2, pp. 1827-1830, Oct. 2000.
[7] C. T. Hsu and J. L. Wu, “Hidden digital watermarks in images,” IEEE Trans. on Image Processing, vol. 8, no. 1, Jan. 1999.
[8] L. Y. Liang and W. Gao, “General schemes for blind watermark,” IEEE International Conference on Robotics, Intelligent Systems and Signal Processing, vol. 2, pp. 1185-1190, Oct. 2003.
[9] Y. L. Liu, W. Gao, M. H. Cui and Y. X. Song, “General blind watermark schemes,” International Conference on Web Delivering of Music, pp. 143-149, 2002.
[10] Y. Choi and K. Aizawa, “Digital watermarking using inter-block correlation,” International Conf. on Image Processing, vol. 2, pp.216-220, October 1999.
[11] Y. Choi and K. Aizawa, “Digital watermarking technique using block correlation of DCT coefficients,” Electronics and Communications in Japan, vol. 85, Issue 9, pp. 23-31, Aug. 2002.
[12] Y. L. Wang and A. Pearmain, “Blind image data hiding based on self reference,” Pattern Recognition Letters 25, pp. 1681-1689, Aug. 2004.
[13] I. J. Cox, M. L. Miller and A.L. McKellips, “Watermarking as communications with side information,” Proceedings of the IEEE, vol. 87, issue 7, pp. 1127-1141, July 1999.
[14] C. S. Lu, H. Y. M. Liao and M. Kutter, “A new watermarking scheme resistant to denoising and copy attacks,” IEEE Fourth Workshop on Multimedia Signal Processing, pp. 505-510, Oct. 2001.
[15] C. S. Lu, H. Y. M. Liao and M. Kutter, “Denoising and copy attacks resilient watermarking by exploiting prior knowledge at detector,” IEEE Transactions on Image Processing, vol. 11, Issue 3, pp. 280-292, March 2002.
[16] L. F. Turner, “Digital data security system,” Patent IPN WO 89/08915, 1989.
[17] R. G. van Schyndel, A. Z. Tirkel and C. F. Osborne, “A digital watermark,” in Int. Conf. Image Processing, vol. 2, pp. 86-90, 1994.
[18] J. Brassil, S. Low, N. Maxemchuk and L. O'Gorman, “Electronic marking and identification techniques to discourage document copying,” in Proc. Infocom, pp. 1278-1287, 1994.
[19] S. S. Bedi and S. Verma, “A design of secure watermarking scheme for images in spatial domain,” Annual India Conference, pp. 1-6, Sept. 2006.
[20] I. J. Cox, J. Kilian, F. T. Leighton and T. Shamoon, “Secure spread spectrum watermarking for multimedia,” IEEE Trans. Image Processing, vol. 6, no. 12, pp. 1673-1687, Dec. 1997.
[21] M. Barrni, F. Bartolini and V. Cappellini, “Copyright protection of digital images by embedded unperceivable marks,” Image and vision computing, vol. 16, pp. 897-906, 1998.
[22] W. N. Lie, G. S. Lin, C. L. Wu and T. C. Wang, “Robust image watermarking on the DCT domain,” IEEE international symposium on circuits and systems, vol. 1, pp. 228-231, May 2000.
[23] W. Zhu, Z. Xiong and Y. Q. Zhang, “Multiresolution watermarking for images and video,” IEEE Trans. Circuits system video technol., vol. 9, no. 4, pp. 545-550, June 1999.
[24] M. Tsai, K. Yuang and Y. Chen, “Joint wavelet and spatial transformation for digital watermarking,” IEEE Trans. on Consumer Electronics, vol. 46, issue 1, pp. 241-245, Feb. 2000.
[25] X. G. Xia, C. G. Boncelet and G. R. Arce, “Wavelet transform based watermark for digital images,” Optics Express, vol. 3, issue 12, pp. 497-511, 1998.
[26] D. Kundur and D. Hatzinakos, “A robust digital image watermarking method using wavelet-based funsion,” IEEE Int. conf. Image Processing, vol. 1, pp. 544-547, 1997.
[27] H. Liu, J. Liu, J. Huang, D. Huang and Y. Q. Shi, “A robust DWT-based blind data hiding algorithm,” IEEE international symposium on circuits and system, vol. 2, pp. 672-675, May 2002.
[28] I. R. Ismaeil and R. K. Ward, “Removal of DCT blocking artifacts using DC and AC filtering,” Conference on Communications, Computers and signal Processing, vol. 1, pp. 229-232, Aug. 2003.
[29] A. Feuer and E. Weinstein, “Convergence analysis of LMS filters with uncorrelated Gaussian data,” IEEE Trans. Acoust., Speech, Signal Processing, vol. 33, no. 1, pp. 222–229, 1985.
[30] B. Widrow, J. Mccool and M. Ball, “The complex LMS algorithm,” Proc. IEEE, vol. 55, pp. 719–720, 1974.
[31] T. Aboulnasr and K. Mayyas, “A robust variable step-size LMS-type algorithm: analysis and simulations,” IEEE Trans. Signal Processing, vol. 45, no. 3, pp. 631–639, Mar. 1997.
[32] R. H. Kwong and E.W. Johnston, “A variable step size LMS algorithm,” IEEE Trans. Signal Processing, vol. 40, no. 7, pp. 1633–1642, Jul. 1992.
[33] 戴顯權，《資料壓縮》，第二版，高雄，紳藍，民國九十一年。