Neural network – based zero - watermarking system for digital images
Keywords:
Digital watermark, deep neural network, intellectual property protection, image security.Abstract
Image authentication has become a paramount necessity in response to the rampant transfer of digital information. In this context, this paper presents an approach for protection of distortion-free digital image. The proposed method is based on extracting a feature map using a deep neural network, which consists of two branches, one for watermark classification and another for feature extraction. This neural network model shares weights between the branches since the watermark maintains image features, thus increasing the system's efficiency. The use of a neural model for feature selection allows the generation of image-specific feature learning, detecting them regardless of whether the image has been distorted. Subsequently, the feature map is combined with the user's watermark, generated from image edge detection, creating a unique stego-image. To evaluate the robustness of the generated watermark, different tests were performed on different types of geometric attacks, such as translation, rotation, scaling, and cropping, as well as advanced image processing, such as JPEG compression, filtering, and noise addition. Three main metrics were used to evaluate the efficiency: Bit Error Rate (BER), Normalized Cross-Correlation (NCC), and Structural Similarity Index (SSIM). The results demonstrate that the proposed method exhibits remarkable robustness to a variety of attacks, reflected in the values of the average BER of 0.0059, a high NCC of 0.9936, as well as an SSIM of 0.9195. These results highlight the effectiveness and reliability of the proposed approach in digital images protection and authentication.
References
[1] W. Wan, J. Wang, Y. Zhang, J. Li, H. Yu y J. Sun, “A comprehensive survey on robust image watermarking”, Neurocomputing, vol. 488, 2022, pp. 22-24, doi: https://doi.org/10.1016/j.neucom.2022.02.083.
[2] P. Aparna y P. V. V. Kishore, “A Blind Medical Image Watermarking for Secure E-Healthcare Application Using Crypto-Watermarking System”, Journal of Intelligent Systems, vol. 29, nº 1, pp. 1558-1575, 2020, doi: https://doi.org/10.1515/jisys-2018-0370.
[3] M. Magdy, N. Ghali, S. Ghoniemy y K. Hosny, “Multiple Zero-Watermarking of Medical Images for Internet of Medical Things”, IEEE Access, vol. 10, pp. 38821-38831, 2022, doi: https://doi.org/10.1109/ACCESS.2022.3165813.
[4] O. P. Singh, K. Singh y G. Srivasta, “Image watermarking using soft computing techniques”, Multimedia Tools and Applications, 2020, pp. 30367-30398, doi: https://doi.org/10.1007/s11042-020-09606-x.
[5] S. Xing, Z. Cheng, C. Y. Ji, J. Chen y L. Qi, “Remote Sensing Image Zero-Watermarking Algorithm Based on Bemd”, Journal of Physics: Conference Series, vol. 1865, 2021, doi: https://doi.org/10.1088/1742-6596/1865/4/042035.
[6] M. Abdullad, A. Ismail y A. Abubakar, "Imperceptibility Analysis for Watermaking Technique Based on Image Block Division Scheme," International Multi-Conference on Systems, Signals & Devices, Tunasia, 2021, doi: https://doi.org/10.1109/SSD52085.2021.9429305.
[7] N. Jimson y K. Hemachandran, "DFT Based Coefficient Exchange Digital Image Watermarking," in Conference on Intelligent 9omputing and Control Systems, Madurai, India, 2018, doi: https://doi.org/10.1109/ICCONS.2018.8663122.
[8] Y. Guanghui y Q. Hao, "Digital watermarking secure scheme for remote sensing image protection," China Communications, vol. 17, no. 4, pp. 88-98, 2020, doi: https://doi.org/10.23919/JCC.2020.04.009.
[9] Z. Yuan, D. Liu, X. Zhang, H. Wang y Q. Su, “DCT-based color digital image blind watermarking method with variable steps”, Multimedia Tools and Applications, vol. 2020 n° 79, pp. 30557–30581, 2020, doi: https://doi.org/10.1007/s11042-020-09499-w.
[10] B. Ram, “Digital Image Watermarking Technique Using Discrete Wavelet Transform and Discrete Cosine Transform”. SSRN, doi: http://dx.doi.org/10.2139/ssrn.4173742.
[11] M. Veni y T. Meyyappan, “Digital image Watermark embedding and extraction using oppositional fruit Fly algorithm”, Multimedia Tools and Applications, vol. (2019), n°78, pp. 27491–27510, 2019, doi: https://doi.org/10.1007/s11042-019-7650-0.
[12] L. Jing, Z. Sun, K. Chen, X. Wen y X. Cheng, “Remote Sensing Image Zero Watermarking Algorithm Based on DFT”, Journal of Physics: Conference Series, vol. 16865, 2021, doi: https://doi.org/10.1088/1742-6596/1865/4/042034.
[13] S. Xing, T. Y. Li y J. Liang, “A Zero-Watermark Hybrid Algorithm for Remote Sensing Image Based on DCT and DFT”, Journal of Physics: Conference Series, vol. 1952, 2021, doi: https://doi.org/10.1088/1742-6596/1952/2/022049.
[14] Y Fang, J. Liu, J. Li1, J. Cheng, J. Hu, D. Yi, X. Xia y U. A. Bhatti, “Robust zero-watermarking algorithm for medical images based on SIFT and Bandelet-DCT”, Multimedia Tools and Applications, vol. (2022), n° 81, pp. 16863–16879, doi: https://doi.org/10.1007/s11042-022-12592-x.
[15] D. Li, Y. Chen, J. Li, L. Cao, U. A. Bhatti y P. Zhang, “Robust watermarking algorithm for medical images based on accelerated-KAZE discrete cosine transform”, IET Biometrics, 2022, doi: https://doi.org/10.1049/bme2.12102.
[16] Q. Zhang, J. Lu y Y. Jin, “Artificial intelligence in recommender systems”, Complex & Intelligent Systems volume, vol. 7, 2021, p. 439–457, doi: https://doi.org/ 10.1007/s40747-020-00212-w.
[17] I. Hamamoto y M. Kawamura, “Image Watermarking Technique Using Embedder and Extractor Neural Networks”, IEICE Transactions on Information and Systems, Special Section on Enriched Multimedia - Making Multimedia More Convenient and Safer, vol. E102.D, nº 1, 2019, pp. 19-30, doi: https://doi.org/10.157/transinf.2018MUP0006.
[18] B. Han, J. Du, Y. Jia y H. Zhu, “Zer-Watermarking Algorithm for Medical Image Based on VGG19 Deep Convolutional Neural Network”, Journal of Healthcare Engineering, vol. 2021, doi: https://doi.org/10.1155/2021/5551520.
[19] C. Gong, J. Liu, M. Gong, J. Li, U. A. Bhatti y J. Ma, “Robust medical zero‐watermarking algorithm based on Residual DenseNet”, IET Biometrics, 2022, doi: https://doi.org/10.1049/bme2.12100.
[20] S. A. Nawaz, J. Li, M. U. Shoukat, U. A. Bhatti y M. A. Raza, “Hybrid medical image zero watermarking via discrete wavelet transform-ResNet101 and discrete cosine transform”, Computers and Electrical Engineering, n°112, vol. 108985, 2023, doi: https://doi.org/10.1016/j.compeleceng.2023.108985.
[21] T. Huang, J. Xu, S. Tu y B. Han, “Robust zero-watermarking scheme based on a depthwise overparameterized VGG network in healthcare information security”, Biomedical Signal Processing and Control, n° 81, vol. 104478, 2023, doi: https://doi.org/10.1016/j.bspc.2022.104478.
[22] D. Li, J. L, U. A. Bhatti, S. A. Nawaz, J. Liu, Y. W. Chen y L. Cao, “Hybrid Encrypted Watermarking Algorithm for Medical Images Based on DCT and Improved DarkNet53”, Electronics, n° 12, vol. 1554, 2023, doi: https://doi.org/10.3390/electronics12071554.
[23] F. Dong, J. Li, U. A. Bhatti, J. Liu, Y. W. Chen y D. Li, “Robust ZeroWatermarking Algorithm for Medical Images Based on Improved NasNet-Mobile and DCT”, Electronics, n° 12, vol. 3444, 2023, doi: https://doi.org/10.3390/electronics12163444
[24] A. N. Fierro, M. Nakano, M. Cedillo-Hernandez, L. Cleofas-Sanchez y H. M. Pérez, “A Robust Image Zero-watermarking using Convolutional Neurañ Networks, 2019 7th International Workshop on Biometrics and Forensics (IWBF), Cancun, Mexico, 2019, doi: 10.1109/IWBF.2019.8739245.
[25] X, Zhong, P-C. Huang, S. Mastorakis y F. Y. Shih, “Automatend and Robust Image Watermarking Scheme Based on Deep Neural Networks”, IEEE Transactions on Multimedia, vol 23, 2021, doi: 10.1109/TMM.2020.3006415.
[26] Y. Quan, H. Teng, Y. Chen y H. Ji, “Watermarking Deep Neural Networks in Image Processing”, IEEE Transactions on Neural Networks and Learning Systems, vol 32, n° 5, 2021, doi: 10.1109/TNNLS.2020.2991378.
[27] J. Fei, Z. Xia, B. Tondi, M. Barni, “Supervised GAN Watermarking for Intellectual Property Protection”.
[28] “Kaggle”. Available: https://www.kaggle.com/datasets/prasunroy/natural-images?resource=download. [Último acceso: 25 01 2023], Roy, Prasun, Ghoshm Subhankar Bhattacharya, Saumik and Pal,” Effects of Degradations on Deep Neural Network Architectures”, preprint arXiv:1807.10108, 2018.
[29] “Medical Segmentation Decathlon”, [En línea]. Available: http://medicaldecathlon.com/. [Último acceso: 24 07 2023]. Amber L. Simpson, Michela Antonelli, Spyridon Bakas, Michel Bilello, Keyvan Farahani, Bram van Ginneken, Annette Kopp-Schneider, Bennett A. Landman, Geert Litjens, Bjoern Menze, Olaf Ronneberger, Ronald M. Summers, Patrick Bilic, Patrick F. Christ, Richard K. G. Do, Marc Gollub, Jennifer Golia-Pernicka, Stephan H. Heckers, William R. Jarnagin, Maureen K. McHugo, Sandy Napel, Eugene Vorontsov, Lena Maier-Hein, M. Jorge Cardoso, “A large annotated medical image dataset for the development and evaluation of segmentation algorithm, 2019, doi: https://doi.org/10.48550/arXiv.1902.09063
[30] R. P. Naik, U. S. Acharya, S. Lal y P. Krishnan “Performance investigation of underwater wireless optical system for image transmission through the oceanic turbulent optical medium”, Optical and Quantum Electronics, vol. 54, n° 251, 2022, doi https://doi.org/10.1007/s11082-022-03611-0.
[31] L. Yu, S. Feng, B. Liang y X. Chen, "High-precision Solar Image Registration Using Normalized Cross-correlation and Intensity", 2021 40th Chinese Control Conference (CCC), Shanghai, China, pp. 3253-3257, 2021, doi: 10.23919/CCC52363.2021.9550094.
[32] A.N. Omara, Tarek M. Salem, Sherif Elsanadily, M.M. Elsherbini, “SSIM-based sparse image restoration”, Journal of King Saud University - Computer and Information Sciences, vol. 34, n° 8, Part B, pp. 6243-62542022, doi: https://doi.org/10.1016/j.jksuci.2021.07.024.
[33] M. Fullan, J. Quinn y J. McEachen, “Deep learning engage the world change the world”, Thousands Oaks, California: Corwin, 2018.
[34] M. Ekman, “Learning deep learning: Theory and practice of neural networks, computer vision, natural language processing, and transformers using tensorflow”, Boston: Nvidia, 2022.
[35] A. Krizhevsky, I. Sutskever, G. Hinton, “Imagenet classification with deep convolutional neural networks,” Communications of the ACM, vol. 60, nº 6, 2017, p. 84–90, doi: https://doi.org/10.1145/3065386.
[36] A. N. Fierro, M. Nakano, H. M. Pérez y K. Yanai, “Redes Convolucionales Siamesas y Tripletas para la Recuperación de Imágenes Similares en Contenido”, Información tecnológica, 2019, doi: http://dx.doi.org/10.4067/S0718-07642019000600243.
