Abstract
This paper proposes a new method for copyright protection of deep neural networks designed for solving image classification tasks. The main idea of the method is to embed digital watermarks into the deep model by fine-tuning on a unique set of images, called triggers, represented in the form of pseudo-holographic signals (pseudo-holograms). A pseudo-hologram is a two-dimensional sinusoidal signal that encodes a binary sequence of arbitrary length. By changing the phase of each sinusoid, it is possible to form various pseudo-holograms encoding the same binary sequence. The proposed watermarking method consists in construction of a training set by producing a required number of pseudo-holograms on the basis of binary sequences, which are unique for each class. Thus, the class label assigned to each pseudo-hologram depend on the sequence encoded in it. The procedure of watermark verification is performed by sending various random pseudo-holograms as model input and evaluating the accuracy of classification. High rate of successful predictions indicates that input images are constructed based on the identification key of the legal owner. Experimental studies confirm the efficiency of the method for various model architectures and prove the compliance with all quality criteria required for the methods of deep model watermarking.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Uchida, Y., Nagai, Y., Sakazawa, S., Satoh, S.: Embedding watermarks into deep neural networks. In: Proceedings of the 2017 ACM on International Conference on Multimedia Retrieval, pp. 269–277 (2017)
Fan, L., Ng, K.W., Chan, C.S.: Rethinking deep neural network ownership verification: Embedding passports to defeat ambiguity attacks. In: Proceedings of the Advances in Neural Information Processing Systems, pp. 4714–4723 (2019)
Wang, T., Kerschbaum, F.: Robust and undetectable white-box watermarks for deep neural networks. arXiv: 1910.14268 (2019)
Nagai, Y., Uchida, Y., Sakazawa, S., Satoh, S.: Digital watermarking for deep neural networks. Int. J. Multimedia Inf. Retrieval 7(1), 3–16 (2018). https://doi.org/10.1007/s13735-018-0147-1
Chen, H., Darvish Rohani, B., Koushanfar, F.: DeepMarks: A digital fingerprinting framework for deep neural networks. In: Proceedings of the 2019 on International Conference on Multimedia Retrieval (ICMR 2019), pp. 105–113 (2019)
Wang, J., Wu, H., Zhang, X., Yao, Y.: Watermarking in deep neural networks via error back-propagation. Electron. Imaging 2020(4), 221–229 (2020)
Kuribayashi, M., Tanaka, T., Suzuki, S., Yasui, T., Funabiki, N.: White-box watermarking scheme for fully-connected layers in fine-tuning model. In: Proceedings of the 2021 ACM Workshop on Information Hiding and Multimedia Security, pp. 165–170 (2021)
Wang, T., Kerschbaum, F.: RIGA: covert and robust white-box watermarking of deep neural networks. In: Proceedings of the Web Conference 2021, pp. 993–1004 (2021)
Botta, M., Cavagnino, D., Esposito, R.: NeuNAC: a novel fragile watermarking algorithm for integrity protection of neural networks. Inf. Sci. 576, 228–241 (2021)
Le Merrer, E., Pérez, P., Trédan, G.: Adversarial frontier stitching for remote neural network watermarking. Neural Comput. Appl. 32(13), 9233–9244 (2019). https://doi.org/10.1007/s00521-019-04434-z
Adi, Y., Baum, C., Cisse, M., Pinkas, B., Keshet, J.: Turning your weakness into a strength: watermarking deep neural networks by backdooring. In: Proceedings of the 27th USENIX Security Symposium (USENIX Security 2018), pp. 1615–1631 (2018)
Deeba, F., Tefera, G., She, K., Memon, H.: Protecting the intellectual properties of digital watermark using deep neural network. In: Proceedings of the 2019 4th International Conference on Information Systems Engineering (ICISE), pp. 91–95 (2019)
Zhang, J., Gu, Z., Jang, J., Wu, H., Stoecklin, M.P., Huang, H., Molloy, I.: Protecting intellectual property of deep neural networks with watermarking. In: Proceedings of the 2018 on Asia Conference on Computer and Communications Security, pp. 159–172 (2018)
Sakazawa, S., Myodo, E., Tasaka, K., Yanagihara, H.: Visual decoding of hidden watermark in trained deep neural network. In: 2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR), pp. 371–374 (2019)
Wang, G., Chen, X., Xu, C.: Adversarial watermarking to attack deep neural networks. In: ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1962–1966 (2019)
Guo, J., Potkonjak, M.: Watermarking deep neural networks for embedded systems. In: Proceedings of the 2018 IEEE/ACM International Conference on Computer-Aided Design (ICCAD), pp. 1–8 (2018)
Jebreel, N. M., Domingo-Ferrer, J., Sánchez, D., Blanco-Justicia, A.: KeyNet: an asymmetric key-style framework for watermarking deep learning models. Appl. Sci. 11 (2021). https://doi.org/10.3390/app11030999
Li, Z., Hu, C., Zhang, Y., Guo, S.: How to prove your model belongs to you: a blind-watermark based framework to protect intellectual property of DNN. In: Proceedings of the 35th Annual Computer Security Applications Conference, pp. 126–137 (2019)
Namba, R., Sakuma, J.: Robust watermarking of neural network with exponential weighting. In: Proceedings of the 2019 ACM Asia Conference on Computer and Communications Security, pp. 228–240 (2019)
Rouhani, B.D., Chen, H., Koushanfar, F.: Deepsigns: a generic watermarking framework for ip protection of deep learning models. arXiv:1804.00750 (2018)
Zhang, Y.-Q., Jia, Y.-R., Niu, Q., Chen, N.-D.: DeepTrigger: a watermarking scheme of deep learning models based on chaotic automatic data annotation. IEEE Access 8, 213296–213305 (2020)
Zhong, Q., Zhang, L.Y., Zhang, J., Gao, L., Xiang, Y.: Protecting IP of deep neural networks with watermarking: a new label helps. In: Lauw, H.W., Wong, R.-W., Ntoulas, A., Lim, E.-P., Ng, S.-K., Pan, S.J. (eds.) PAKDD 2020. LNCS (LNAI), vol. 12085, pp. 462–474. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-47436-2_35
Xu, X., Li, Y., Yuan, C.: “Identity bracelets” for deep neural networks. IEEE Access 8, 102065–102074 (2020)
Zhao, J., Hu, Q., Liu, G., Ma, X., Chen, F., Hassan, M.: AFA: adversarial fingerprinting authentication for deep neural networks. Comput. Commun. 150, 488–497 (2020)
Cao, X., Jia, J., Gong, N.Z.: IPGuard: Protecting the intellectual property of deep neural networks via fingerprinting the classification boundary. In: Proceedings of the 2021 ACM Asia Conference on Computer and Communications Security (ASIA CCS 2021), pp. 14–25 (2021)
Kim, W., Lee, K.: Digital watermarking for protecting audio classification datasets. In: ICASSP 2020–2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 2842–2846 (2020)
Chen, H., Zhang, W., Liu, K. Chen, K., Fang, H., Yu, N.: Speech pattern based black-box model watermarking for automatic speech recognition. In: Proceedings of the 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 3059–3063 (2022)
Wang, Y., Wu, H.: Protecting the intellectual property of speaker recognition model by black-box watermarking in the frequency domain. Symmetry 14(3), 619 (2022)
Wu, H., Liu, G., Yao, Y., Zhang, X.: Watermarking neural networks with watermarked images. IEEE Trans. Circuits Syst. Video Technol. 31(7), 2591–2601 (2021)
Zhang, J., Chen, D., Liao, J., Zhang, W., Feng, H., Yu, N.: Deep model intellectual property protection via deep watermarking. IEEE Trans. Pattern Anal. Mach. Intell. 44, 4005–4020 (2021)
Quan, Y., Teng, H., Chen, Y., Ji, H.: Watermarking deep neural networks in image processing. IEEE Trans. Neural Netw. Learn. Syst. 32(5), 1852–1865 (2021)
Chen, K., Guo, S., Zhang, T. Li, S., Liu, Y.: Temporal watermarks for deep reinforcement learning models. In: Proceedings of the 20th International Conference on Autonomous Agents and MultiAgent Systems (AAMAS 2021), pp. 314–322 (2021)
Vybornova, Y.: Method for protection of heterogeneous data based on pseudo-holographic watermarks. In: Proceedings of 2021 9th International Symposium on Digital Forensics and Security (ISDFS), pp. 1–5 (2021)
Torchvision models subpackage. https://pytorch.org/vision/stable/models.html
CIFAR10 and CIFAR100 datasets. http://www.cs.toronto.edu/~kriz/cifar.html
Acknowledgments
The reported study was funded by RSF (Russian Science Foundation) grant № 21-71-00106, https://rscf.ru/en/project/21-71-00106/
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Vybornova, Y. (2023). Digital Watermarking Method for Copyright Protection of Deep Neural Networks. In: Arai, K. (eds) Intelligent Computing. SAI 2023. Lecture Notes in Networks and Systems, vol 739. Springer, Cham. https://doi.org/10.1007/978-3-031-37963-5_42
Download citation
DOI: https://doi.org/10.1007/978-3-031-37963-5_42
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-37962-8
Online ISBN: 978-3-031-37963-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)