為了方便影像在網路傳送、監視系統及醫學影像各方面之使用，
對於原始影像加以使用影像處理及失真性壓縮等運算程序，近來發展出一種可忍受一般影像處理之操作但對於惡意竄改之區域能敏感偵測之半易碎數位浮水印技術。在半易碎浮水印之技術上，利用奇異值分解(Singular Value Decomposition, SVD)後之奇異值來嵌入浮水印是一可行之方法。但修改奇異值大都會影響到影像之品質並降低浮水印之強健性，如何改善此現象是一項值得深入研究的議題。
本論文利用影像次取樣(Subsampling)產生的子影像(Subimage)彼此具有高相似度的特性，提出了一個次取樣影像基於奇異值分解
(SVD)之半易碎數位浮水印技術，此方法是以次取樣方式將原始影像
分成四個相似之子影像，利用其中兩個子影像產生之區塊特徵相關値再加上調整關係之一相關因子(Correlation Factor, CF)來調整浮水印的易碎性與強健性。而對於作為保護著作權資訊之浮水印則嵌入於另外兩個子影像區塊奇異值分解後之係數中。利用子影像區塊之關聯性，影像受到竄改即能偵測出其受到竄改的區域，同時亦能萃取出著作權資訊之浮水印。
經由實驗顯示，本論文所提出之方法，除了在萃取浮水印的過程
中不需要原始影像，並證明本論文所利用之次取樣影像配合作為奇異值分解之半易碎浮水印嵌入技術，較之於其他方法，對於高斯雜訊和銳化等攻擊法可保有較佳之強健性，而其易碎之特性則可以準確定位竄改之位置。

The original images are usually handled by general image
manipulations or loss compression for some kinds of situations such as the Internet, surveillance system and medicine. Semi-fragile digital watermarking techniques allow content-preserving manipulations while are sensitive to malicious tamper. These techniques modify the singular
values through the singular value decomposition (SVD) in recent years. It is a research because the watermarked image quality may be decreased by modifying the singular values.
In this paper, we utilize the high similar property of each subimage through subsampling and propose a subsampling image SVD-based semi-fragile watermarking technique with use of the high similar property
of each subimage through subsampling. First, four subimages are obtained by subsampling. We use two of subimages to generate a block correlation with the correlation factor (CF) for adjusting between the fragility and robustness of the embedded watermark. Then, The watermark are embedded into two SVD coefficients of the other subimages. The proposed technique can detect the modified regions with block unit because we employ the block correlation of all subimage.
Experiment results show that the precision of this technique not only overcomes several attacks such as the gaussian noise attack and sharpening attack, but also detects the tampered regions.

[1] D. Kundur and D. Hatzinakos, “Digital watermarking for telltale tamper proofing
and authentication,” Proc. IEEE vol.87, pp.1167-1180, July 1999.
[2] S. Bhattacharjee and M. Kutter, “Compression tolerant image authentication,” in
Proc. IEEE ICIP, Chicago, IL, Oct. 1998.
[3] Lin, C.-Y. and Chang, S.-F, “A robust image authentication method distinguish
JPEG compression from malicious manipulation,” IEEE Transactions on Circuits
and System for Video Technol., vol.11, pp.153-168, Feb. 2001.
[4] R. G. Van Schyndel, A. Z. Tirkel, and C.F. Osborne, “A digital watermark,” in
Proc. IEEE ICIP, Austin, Texas, vol.2, pp.86-90, Nov. 1994.
[5] R. Wolfgang and E. J. Delp, “A watermark for digital images”, in Proc. IEEE
ICIP, vol.3, pp.219-222, Sept. 1996.
[6] Chun-Shien Lu and Hong-Yuan Mark Liao, “Multipurpose watermarking for
image authentication and protection”, IEEE Trans. Image Processing, vol.10,
pp.1579 - 1592, Oct. 2001.
[7] Ruizhen Liu and Tieniu Tan , “An SVD-based watermarking scheme for
protecting rightful ownership,“ IEEE Trans. on Multimedia, vol.4, pp.121 - 128,
Mar. 2002.
[8] Rui Sun, Hong Sun and Tianren Yao, “A SVD- and quantization based
semi-fragile watermarking technique for image authentication,” IEEE ICSP, vol.
2, pp.1592 - 1295, Aug. 2002.
[9] Yuan Jun, Cui Guo-Hua and Zhang Yi-Jia, “A practical multipurpose color image
watermarking algorithm for copyright protection and image
authentication,“ IEEE ICDT, pp.72 – 72, 2006.
[10] Wai C. Chu, “DCT-based image watermarking using subsampling,” IEEE Trans.
on Multimedia, vol.5, pp.34 – 38, Mar. 2003.