在本論文中提出一套結合空間域與頻率域之數位半色調影像浮水印演算法。首先，在數位灰階影像轉換為數位半色調影像的過程中嵌入浮水印，避免直接從空間域上做修改，影響半色調影像品質與視覺效果；為了提高浮水印系統嵌入容量與增加強健性，我們在具有空間域浮水印之半色調影像中，利用小波轉換找出紋理變化大以及灰階值極大與極小之區域，當作頻率域浮水印嵌入之特定區塊。之後利用更改特定區塊之高頻係數值，嵌入頻率域浮水印。本文利用Donut Masks Zernike Moment求得具有混合型浮水印之半色調影像的旋轉角度與縮放倍率之參數，進而抵抗多重攻擊。實驗結果說明本文所提出之數位浮水印系統可抵抗旋轉、縮放、黑色塗鴉、剪裁、胡椒鹽雜訊以及列印與掃瞄攻擊，具有非常好的強健性。

In this thesis, a hybrid watermarking for digital halftone images was proposed. At first, the spatial watermarking was adopted by the ordered dithering technique which uses two dither cells in the digital halftoning. In view of the shortcomings of the spatial watermarking, the spatial watermarked halftone image was divided into sub-images with 8x8 pixels and transformed by binary pseudo-wavelet transform. The frequency watermarking was embedded into the high frequency components of the designated sub-images. Finally, the donut Masks Zernike moment is adopted to detect rotation and scale of the Hybrid watermarked halftone image. Experimental results show that the hybrid watermarked halftone image with high transparency and the is robust to the malicious attacks such as corrupting, cropping, rotating, scaling, salt & pepper noise. The correct decoding rate is close to 0.9 after the attack of print & scan.

目錄

第一章 序論..................................................1
1.1 數位影像浮水印.....................................1
1.1.1 強健型數位影像浮水印設計之要點................3
1.2 研究動機...........................................5
1.3 論文架構...........................................7
第二章 相關背景知識 ..........................................9
2.1 數位半色調技術 .....................................9
2.1.1 點陣調色法...................................10
2.1.2 誤差擴散法...................................15
2.2 小波轉換..........................................17
2.3 二元擬小波轉換....................................21
2.4 Donut Masks Zernike Moment........................24
2.4.1 Zernike Moment................................25
2.4.2 Zernike動量的旋轉不變性.......................26
2.4.3 Zernike動量的大小不變性 ......................27
2.2.4 Zernike動量的位移不變性 .......................28
2.2.5 Donut Masks Zernike Moment 參數評估............28
2.2.6 DMAD旋轉角度偵測法.........................31
2.2.7 DMSD縮放倍率偵測法 ........................33
第三章 混合型數位浮水印系統之設計...........................35
3.1 混合型數位浮水印嵌入演算法........................37
3.1.1 空間域浮水印嵌入演算法.......................37   
3.1.2 頻率域浮水印嵌入演算法.......................40
3.2 混合型數位浮水印擷取演算法........................44
   3.2.1 空間域浮水印擷取演算法.......................44
   3.2.2 頻率域浮水印擷取演算法.......................45
3.3 空間域與頻率域浮水印演算法之獨立性探討............46
第四章 系統評估.............................................49
4.1 影像品質評估......................................50
4.2 強健性評估........................................51
			4.2.1 錯誤糾正碼－重複編碼.........................52
4.2.2 攻擊測試.....................................53
第五章 結論與未來展望.......................................58
	   5.1 結論..............................................58
		5.2 未來展望..........................................58
參考文獻....................................................60

圖目錄
圖1.1(a)典型浮水印嵌入法流程圖................................3
(b)典型浮水印抽取法流程圖................................3
圖2.1經量化之點陣調色法流程圖...............................12
圖2.2 Lena影像經量化後利用集中式點陣法所得之半色調影像...... 13
圖2.3 Lena影像經量化後利用分散式點陣法所得之半色調影像...... 13
圖2.4未量化之點陣調色法流程圖............................... 13
圖2.5集中式點陣法-未量化
 (a)512 512 Lena 影像 (b)256 256 Lena影像................. 14
圖2.6分散式點陣法-未量化
     (a)512 512 Lena 影像 (b)256 256 Lena影像................. 14
圖2.7誤差擴散法之流程圖 ..................................... 15
圖2.8使用Floyd-Steinberg誤差擴散濾鏡
     (a)512 512 Lena 影像 (b)256 256 Lena影像................. 17
圖2.9使用Jarvis誤差擴散濾鏡
      (a)512 512 Lena 影像 (b)256 256 Lena影像................ 17
圖2.10 ( a )時域訊號  ( b )經傅立葉轉換後的頻域訊號............. 18
圖2.11 短時距傅立葉分析...................................... 19
圖2.12 小波係數之階層關係................................... 21
圖2.13 ㄧ階小波轉換示意圖................................... 21
圖2.14 (a)二元擬小波轉換順向基底 (b)二元擬小波轉換反向基底.... 23
圖2.15 BPWT係數示意圖...................................... 23
圖2.16 (a)原始半色調影像 (b)保留BPWT低頻係數之半色調影像.... 24
圖2.17 影像分成外徑不同的甜甜圈影像(a)~(o) ................... 30
圖2.18  DMAD旋轉角度偵測流程圖............................ 31
圖2.19  DMSD縮放倍率偵測流程圖............................. 33
圖3.1 基於半色調影像區塊之混合型浮水印嵌入系統流程圖........ 36
圖3.2 高頻子頻帶對應之黑白影像與Map S....................... 37
圖3.3 空間域浮水印嵌入流程圖................................ 38
圖3.4 分散式臨界值矩陣C0與C1.............................. 39
圖3.5 (a) 解析度為512 512的灰階影像
(b) 具有空間域強健型浮水印之半色調影像.................. 40
圖3.6 頻率域浮水印嵌入流程圖................................ 40
圖3.7 (a) 標示出強邊緣區域之Map  
      (b) 標示出低灰階值區域之Map  
      (c) 標示出高灰階值區域之Map  ........................ 41
圖3.8 BPWT高頻係數示意圖................................... 41
圖3.9 (a) 具有空間域強健型浮水印之半色調影像
(b) 具有頻率域與空間域強健型浮水印之半色調影像..........43
圖3.10 基於半色調區塊之混合型浮水印擷取系統流程圖............44
圖3.11 解碼區塊圖樣與原始區塊圖樣之相關係數曲線圖............47
圖3.12 各種情形下解碼區塊圖樣與原始區塊圖樣之相關係數曲線圖 47
圖4.1 MPSNR 流程圖......................................... 50
圖4.2 Gaussian 低通濾波器..................................... 50
圖4.3 (a) 原始灰階影像，(512 512) 、
     (b) 嵌入空間域浮水印，(32 32)、
(c) 嵌入頻率域浮水印，(32 32)、
     (d) 具有空間域與頻率域浮水印之半色調影像，(512 512) ...... 52
圖4.4 受到各種攻擊之半色調影像Lena.......................... 55
圖4.5 從圖4.4(a)~(n)中所擷取出之空間域強健型浮水印(經ECC糾錯) .55
圖4.6 從圖4.4(a)~(n)中所擷取出之頻率域強健型浮水印(經ECC糾錯) .55
圖4.7 在圖4.4中(a)~(n)之空間域與頻率域中浮水印，經ECC糾錯後所得到的正確解碼率之比較分析圖............................. 56

表目錄
表2.1 經量化之臨界值矩陣
(a)	集中式矩陣 (b) 分散式矩陣........................... 11
表2.2 未量化之臨界值矩陣
(a) 集中式矩陣 (b) 分散式矩陣........................... 11
表2.3 (a) Floyd-Steinberg誤差擴散濾鏡 
 (b) Jarvis誤差擴散濾鏡.................................. 16
表2.4 (a)XOR邏輯運算子 (b)AND邏輯運算子.................... 21
表3.1 在各種情況下所得到的平均相關係數一覽表................ 48
表4.1兩種強健性浮水印分別嵌入不同容量之MPSNR比較.......... 51
表4.2 不同方法之MPSNR比較................................. 51
表4.3 受到各種攻擊所測得空間域浮水印與頻域浮水印之正確解碼率.57

[1]  W. Bender and D. Gruhl et al.,“Techniques for data hiding ,”IBM System Journal, vol. 35, no. 3&4, pp. 313-336, 1996.
[2]  M. S. Fu and O.C. Au ,“A robust public watermark for halftone images,”Circuits and Systems, 2002. ISCAS 2002. IEEE International Symposium on Vol. 3, 26-29 May 2002, pp.639 – 642. 
[3]  M. S. Fu and O.C. Au ,“A Multi-bit Robust Watermark for Halftone images,”Multimedia and Expo, 2003. ICME '03. Proceedings. 2003 International Conference on Vol. 1, 6-9 July 2003, pp.213-16.
[4]  M. S. Fu and O.C. Au ,“Data hiding in halftone images by stochastic error diffusion,”Acoustics, Speech, and Signal Processing, 2001. Proceedings. (ICASSP '01). 2001 IEEE International Conference on Vol. 3, 7-11 May 2001, pp.1965 – 1968.
[5]  M. S. Fu and O.C. Au ,“A novel Self-Conjugate Halftone Image Watermarking Technique,”Circuits and System, 2003 ISCAS '03. Proceedings of the 2003 International Symposium on Vol. 3, 25-28 May 2003, pp. III-790 - III-793.
[6]  M. S. Fu and O. C. Au ,“A novel method to embed watermark in different halftone images: data hiding by conjugate error diffusion(DHCED),”Multimedia and Expo, 2003. ICME ’03. Proceedings. 2003 International Conference on Vol. 1, 6-9 July 2003, pp, I-609-12
[7]  M. S. Fu and O.C. Au ,“A set of mutually watermarked halftone images,”Image Processing, 2003. ICIP 2003. Proceedings, 2003 International Conference on Vol. 2, 14-17 Sept. 2003, pp.II 467-70.
[8]  M. S. Fu and O.C. Au ,“Correlation-based watermarking for halftone images,”Circuits and Systems, 2004. ISCAS '04. Proceedings of the 2004 International Symposium on Vol. 2, 23-26 May 2004, pp. II - 21-4.
[9]  S. C. Pei and J. M. Guo ,“Novel Data Hiding and Blind Decoding in Dithering Halftone Images,”18th IPPR Conference on Computer Vision, Graphics and Image Processing(CVGIP 2005), 21-23 Aug 2005.
[10] S. C. Pei and J. M. Guo ,“Watermarking in halftone Images with Parity-Matched Error Diffusion,”Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on Vol. 2, March18-23, 2005, pp.825-828.
[11] J. Chen and H. Yao ,“A robust watermarking method based on wavelet and Zernike Transform,”Circuits and System, 2004. ISCAS '04, Proceedings of the 2004 International Symposium on Vol. 2, 23-26 May 2004, pp.II-173-6.
[12] H. Z. Hel-Or ,“Copyright labeling of printed images,”Image Processing, 2000.Proceedings 2000 International Conference on Vol.3, 10-13 Sept 2000, pp.702-705.
[13] H. Luo and Z. M. Lu ,“Halftone Image Watermarking Based on the Binary Pseudo-Wavelet Transform,”IIH-MSP2007,November 2007.
[14] R. W. Floyd and L. Steinberg ,“An adaptive algorithm for spatial gray scale,”Proc ,1976 SID 17(2), pp. 75-77.
[15] J. F. Jarvis and C. S. Roberts ,“A new technique for displaying continuous tone images on a bi-level display,”IEEE Trans. Commun., vol. 24, no. 8, pp. 891-898, 1976. 
[16] S. Pigeon and Y. Bengio ,“Binary Pseudowavelets and Applications to Bilevel Image Processing,”Data Compression Conference 1999, pp. 364-373.
[17] P. Sherry and A. Savakis ,“Improved techniques for watermarking halftone images,”Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP '04). IEEE International Conference on Vol. 5, 17-21 May 2004, pp.V-1005-8.
[18] A. Khotanzad and Y. H. Hong ,“Invariant image recognition by Zernike moments,”Pattern Recognition and Machine Intelligence, IEEE Transactions on Vol.12, Issue 5, May 1990, pp.489-497.
[19] C. H. Teh and R. T. Chin ,“On image analysis by the methods of moments,”Pattern Analysis and Machine Intelligence, IEEE Transactions on Vol. 10, Issue4, July 1988, pp.496-513. 
[20] 葉冠廷 ,“多目的半色調影像浮水印技術之研究,”淡江大學,電機工程學系碩士論文,2006.
[21] M. S. Fu and O. C. Au ,“Data Hiding Watermarking for Halftone Images,”IEEE Transactions on Images Processing ,Vol. 11,No.4 , April 2002, pp.477-84.
[22] R. Y. Li and O. C. Au ,“Halftone Image Data Hiding with Block-Overlapping Parity Check,”ICASSP 2007. IEEE International Conference on Vol. 2, April 2007,pp.193-196.
[23] M. S. Fu and O. C. Au ,“Watermarking technique for color halftone images,”Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP '04). IEEE International Conference on Vol. 3, 17-21 May 2004, pp.iii - 381-4.
[24] S. C. Pei and J. M. Guo ,“Novel robust watermarking technique in dithering halftone images,”Signal Processing Letter, IEEE Vol. 12, Issue 4, Apr 2005, pp.333-336.
[25] D. Kacker and J. P. Allebach ,“Joint halftoning and watermarking,”IEEE Trans. Signal Processing, vol. 51, pp. 1054-1068, no. 4, April 2003.
[26] 陳勇志 ,“網點加密圖像光學擷取比對辨識研究,”國立臺灣師範大學,圖文傳播學系碩士論文,2002.
[27] 蕭佩琪 ,“文件底紋之混合網點數位浮水印技術,”國立臺灣師範大學，圖文傳播學系碩士論文,2004.