本研究提出一種基於穩定擴散模型(Stable Diffusion Model)的強韌浮水印方法，旨在保護數位影像著作權的同時盡可能保留原始影像。利用預訓練的穩定擴散模型進行 DDIM inversion 並結合離散餘弦轉換(DCT)將原始影像之特徵經過區塊式離散餘弦轉換後嵌入浮水印資訊，最後使用相同模型進行反向擴散使其還原成帶有浮水印資訊的影像，實驗結果顯示，本研究能夠在未調整模型的情況下完成浮水印嵌入步驟，且產生的浮水印影像具有一定的強韌性可以對抗不同類型的攻擊。

This study introduces a robust watermarking method based on the Stable Diffusion Model, designed to safeguard digital image copyrights while preserving the original visual content as much as possible. A pretrained Stable Diffusion model is employed to perform DDIM inversion, where watermark information is embedded into image features through block-based. Discrete Cosine Transform (DCT). The same model is then applied via reverse diffusion to reconstruct the final watermarked image. Experimental results demonstrate that the proposed method can embed watermarks with unaltered model parameters and exhibits robustness against a range of common attacks.

目錄
第一章	緒論	1
1.1	研究背景與動機	1
1.2	研究目的	3
1.3	論文架構	3
第二章	背景技術與相關研究	5
2.1	基於CNN的浮水印方法	5
2.2	基於GAN的浮水印方法	7
2.3	基於INN的浮水印方法	9
2.4	基於Transformer的浮水印方法	10
2.5	基於擴散模型的浮水印方法	11
第三章	研究方法	15
3.1	整體架構介紹	15
3.2	前處理	16
3.3	浮水印嵌入方法	17
3.4	浮水印偵測	20
第四章	實驗結果與討論	23
4.1	實驗環境	23
4.2	實驗細節	25
第五章	結論與未來展望	34
5.1	結論	34
5.2	未來展望	34
參考文獻	36



		圖目錄
圖 1、Ouyang等人[8]之研究方法架構圖	6
圖 2、Yu等人[13]的架構圖	8
圖 3、IWN架構圖[18]	9
圖 4、Karki架構圖[23]	11
圖 5、CRoSS架構圖[27]	14
圖 6、CRoSS研究成果[27]	14
圖 7、本研究之浮水印嵌入方法架構圖	16
圖 8、不同步驟對於影像的影響	19
圖 9、各類方法的檢測率折線圖	27
圖 10、不同壓縮率下的JPEG攻擊比較圖	30
圖 11、不同嵌入量之成果比較圖	32

表目錄
表 1、12種攻擊的相關參數設定	24
表 2、 3種資料集中的檢測率比較表	26
表 3、浮水印嵌入方法與3種資料集之成果比較	28
表 4、不同參數下新型攻擊的檢測率統計表	32
表 5、每張影像平均運行時間	33

[1]	J. Ho, A. Jain, and P. Abbeel, “Denoising diffusion probabilistic models,” Advances in Neural Information Processing Systems (NeurIPS), Vancouver, BC, Canada, vol. 33, pp. 6840–6851, Dec. 2020.
[2]	J. Song, C. Meng, and S. Ermon, “Denoising diffusion implicit models,” in Proceedings of the International Conference on Learning Representations (ICLR), Virtual Conference, May. 2021.
[3]	R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer, “High-resolution image synthesis with latent diffusion models,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 10684–10695.
[4]	I. J. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative Adversarial Nets,” in Proceedings of the 27th International Conference on Neural Information Processing Systems (NIPS), Montréal, QC, Canada, Oct. 2020, pp. 2672–2680.
[5]	J. Neyman, and E. S. Pearson, “On the problem of the most efficient tests of statistical hypotheses,” Philosophical Transactions of the Royal Society of London. Series A, Containing Papers of a Mathematical or Physical Character, vol. 231, no. 694–706, pp. 289–337, Feb. 1933.
[6]	P. B. Patnaik, “The Non-Central χ²- and F-Distributions and Their Applications,” Biometrika, vol. 36, no. 1–2, pp. 202–232, Jan. 1949.
[7]	Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to document recognition,” in Proceedings of the IEEE, vol. 86, no. 11, pp. 2278–2324, Nov. 1998.
[8]	J. Ouyang, G. Coatrieux, B. Chen, and H. Shu, “Color image watermarking based on quaternion Fourier transform and improved uniform log-polar mapping,” Computers & Electrical Engineering, vol. 46, pp. 419–432, Aug 2015.
[9]	S. K. Padhi, A. Tiwari, and S. S. Ali, “Deep Learning-Based Dual Watermarking for Image Copyright Protection and Authentication,” IEEE Transactions on Artificial Intelligence, vol. 5, no. 12, Dec. 2024, pp. 6134–6145.
[10]	J. Zhu, R. Kaplan, J. Johnson, and L. Fei-Fei, “HiDDeN: Hiding Data with Deep Networks,” in Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany, pp. 657–672, Sep. 2018.
[11]	S. Baluja, “Hiding Images in Plain Sight: Deep Steganography,” in Advances in Neural Information Processing Systems (NeurIPS), Long Beach, CA, USA, pp. 2069–2079, Dec. 2017.
[12]	M. Tancik, B. Mildenhall, and R. Ng, “StegaStamp: Invisible Hyperlinks in Physical Photographs,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, pp. 2117–2126, Jun.2020.
[13]	H. Yu, X. Luo, Z. Zhao, H. Li, and S. Zhang, “Artificial Fingerprinting for Generative Models: Rooting Deepfake Attribution in Training Data,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pp. 14428–14437, Oct 2021.
[14]	N. Lukas, Y. E. Bozkir, and F. Kerschbaum, “Pivotal Tuning Watermarking for Pre-Trained Image Generators,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, pp. 15037–15047, Jun. 2024.
[15]	K. A. Zhang, A. Cuesta-Infante, L. Xu, and K. Veeramachaneni, “SteganoGAN: High Capacity Image Steganography with GANs,” arXiv preprint arXiv:1901.03892, Jan. 2019.
[16]	C. Yu, “Attention Based Data Hiding with Generative Adversarial Networks,” in Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), New York, NY, USA, vol. 34, no. 1, pp. 1120–1128, Feb. 2020.
[17]	J. Yang, K. Liu, X. Kang, E. K. Wong, and Y.-Q. Shi, “Spatial Image Steganography Based on Generative Adversarial Network,” arXiv preprint arXiv:1804.07939, Apr. 2018.
[18]	H.-B. Xu, R. Wang, J. Wei, and S.-P. Lu, “A compact neural network-based algorithm for robust image watermarking,” arXiv preprint arXiv:2112.13491, Dec. 2021.
[19]	J. Jing, X. Deng, M. Xu, J. Wang, and Z. Guan, “HiNet: Deep image hiding by invertible network,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Virtual Event, pp. 4713–4722, Oct. 2021.
[20]	S.-P. Lu, R. Wang, T. Zhong, and P. L. Rosin, “Large-capacity image steganography based on invertible neural networks,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Virtual Event, pp. 10811–108820, Jun. 2021.
[21]	Y. Zhang, H. Chen, and L. Li, “FIIH: Fully invertible image hiding for secure and robust communication,” arXiv preprint arXiv:2407.17155, 2024.
[22]	W. Sun, J. Liu, W. Dong, L. Chen, and K. Niu, “Watermarking for neural radiation fields by invertible neural network,” arXiv preprint arXiv:2312.02456, Dec. 2023.
[23]	B. Karki, C.-H. Tsai, P.-C. Huang, and X. Zhong, “Deep Learning-Based Text-in-Image Watermarking,” in Proceedings of the IEEE 7th International Conference on Multimedia Information Processing and Retrieval (MIPR), San Jose, CA, USA, pp. 376–382, Aug. 2024.
[24]	X. Ke, H. Wu, and W. Guo, “StegFormer: Rebuilding the Glory of Autoencoder-Based Steganography,” in Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), Vancouver, BC, Canada, vol. 38, pp. 2723–2731, Feb 2024.
[25]	B. Wang, Z. Song, and Y. Wu, “Robust Blind Watermarking Framework for Hybrid Networks Combining CNN and Transformer,” in Proceedings of the Asian Conference on Machine Learning (ACML). Istanbul, Turkey, vol. 222, pp. 1417–1432, Nov. 2023.
[26]	Y. Dong, P. Wei, R. Wang, B. Song, T. Wei, and W. Zhou, “Hiding image with inception transformer,” IET Image Processing, vol. 18, pp. 3961–3975, Nov. 2024. 
[27]	J. Yu, X. Zhang, Y. Xu, and J. Zhang, “CRoSS: Diffusion Model Makes Controllable, Robust and Secure Image Steganography,” in Proceedings of the Conference on Neural Information Processing Systems (NeurIPS), New Orleans, LA, USA, Dec. 2023, pp. 1–15. 
[28]	H. Ci, P. Yang, Y. Song, and M. Z. Shou, “RingID: Rethinking Tree-Ring Watermarking for Enhanced Multi-Key Identification,” in Proceedings of the European Conference on Computer Vision (ECCV), Milan, Italy, Sep. 2024, Lecture Notes in Computer Science, vol. 15115, pp. 3–20.
[29]	Z. Yang, K. Zeng, K. Chen, H. Fang, W. Zhang, and N. Yu, “Gaussian Shading: Provable Performance-Lossless Image Watermarking for Diffusion Models,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, Jun. 2024, pp. 12162–12171.
[30]	A. Rezaei, M. Akbari, S. R. Alvar, A. Fatemi, and Y. Zhang, “LaWa: Using Latent Space for In-Generation Image Watermarking,” in Proceedings of the European Conference on Computer Vision (ECCV), Milan, Italy, Sep. 2024, Lecture Notes in Computer Science, vol. 15147, pp. 118–136, 2024.
[31]	D. P. Kingma, and M. Welling, “Auto-encoding variational Bayes,” in Proceedings of the 2nd International Conference on Learning Representations (ICLR), Banff, Canada, Apr. 2014.
[32]	National Institute of Standards and Technology, “FIPS 202: SHA-3 Standard: Permutation-Based Hash and Extendable-Output Functions,” Gaithersburg, MD, U.S.A., Aug. 2015.
[33]	T.-Y. Lin, M. Maire, S. Belongie, L. Bourdev, R. Girshick, J. Hays, P. Perona, D. Ramanan, C. L. Zitnick, and P. Dollár, “Microsoft COCO: Common Objects in Context,” in Proceedings of the European Conference on Computer Vision (ECCV), Zürich, Switzerland, Sep. 2014, Lecture Notes in Computer Science, vol. 8693, pp. 740–755.
[34]	Z. J. Wang, E. Montoya, D. Munechika, H. Yang, B. Hoover, and D. H. Chau, “DiffusionDB: A Large-Scale Prompt Gallery Dataset for Text-to-Image Generative Models,” in Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL), Toronto, Canada, Jul. 2023.
[35]	F. Phillips, and B. Mackintosh, “Wiki Art Gallery, Inc.: A Case for Critical Thinking,” Issues in Accounting Education, vol. 26, no. 3, pp. 593–608, Aug. 2011. 
[36]	J. Ballé, D. Minnen, S. Singh, S. J. Hwang, and N. Johnston, “Variational image compression with a scale hyperprior,” in Proceedings of the International Conference on Learning Representations (ICLR), Vancouver, Canada, Apr. 2018
[37]	Z. Cheng, H. Sun, M. Takeuchi, and J. Katto, "Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modules," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, Jun. 2020.
[38]	X. Zhao, K. Zhang, Z. Su, S. Vasan, I. Grishchenko, C. Kruegel, G. Vigna, Y.-X. Wang, and L. Li, “Invisible Image Watermarks Are Provably Removable Using Generative AI,” in Proceedings of the 38th Conference on Neural Information Processing Systems, 2024
[39]	L. Zhang, X. Liu, A. V. Martin, C. X. Bearfield, Y. Brun, and H. Guan, “Attack-resilient image watermarking using stable diffusion,” in Proceedings of the 38th Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, Dec. 2024. 
[40]	X. Kang, Y. Chen, F. Zhao, and G. Lin, “Multi-dimensional particle swarm optimization for robust blind image watermarking using intertwining logistic map and hybrid domain,” Soft Computing, vol. 24, no. 14, pp. 10561–10584, Jul. 2020.
[41]	D. C. Liu, and J. Nocedal, “On the limited memory BFGS method for large scale optimization,” Mathematical Programming, vol. 45, pp. 503–528, Dec.  1989. 
[42]	M. Radouane, N. I. Zouggari, A. Amraoui, and M. Amraoui, “A robust watermarking scheme based on DCT, IWT and SVD with optimal block,” Journal of Theoretical and Applied Information Technology, vol. 100, no. 11, Nov. 2023.