Robust zero-watermarking algorithm for image authentication using hybrid spatial-frequency features and siamese neural networks
DOI:
https://doi.org/10.29059/cienciauat.v20i1.1983Keywords:
zero-watermarking, deep learning, siamese neural network, ownership authentication, image securityAbstract
The use and distribution of digital files has increased due to advancements in information technologies. This has created the need for the development of copyright protection systems. In the context of digital images, it is essential to minimize security risks associated with unauthorized distribution and ensure the integrity of visual information. The objective of this paper was to develop a robust zero-watermarking algorithm designed for user ownership verification and the recovery of original images in grayscales in cases of manipulation. The proposed method employed a Siamese neural network with a architecture model consisted of two branches: one learns frequency-domain features from discrete wavelet transform coefficients, and other that extract spatial features from the other image. Additionally, a neural network trained with spatial features was used to reconstruct a grayscale version of the original image after tampering. The proposed method proved its using natural color images demonstrated the high effectiveness in processing times, precision and of the watermark recovery process for verifying user ownership and accurately authenticating manipulated images. Highlighted its improvements robustness when encountering geometrical distortions such as rotation, translation, transformation, affine, clipping and scaling, as well as the combination of some distortions. The trained autoencoder preserves high fidelity in grayscale image reconstruction that has suffered alterations by combining manipulations with other distortions. The proposed algorithm proved to be an effective solution for the authentication and protection of copyright in digital images.
References
Amini, M., Sadreazami, H., Ahmad, M. O., & Swamy, M. N. S. (2018). A channel-dependent statistical watermark detector for color images. IEEE Transactions on Multimedia, 21(1), 65-73. https://doi.org/10.1109/TMM.2018.2851447. DOI: https://doi.org/10.1109/TMM.2018.2851447
Amerini, I., Ballan, L., Caldelli, R., Del-Bimbo, A., & Serra, G. (2011). A SIFT-based forensic method for copy-move attack detection and transformation recovery. IEEE Transactions on Information Forensics and Security, 6(3), 1099-1110. https://doi.org/10.1109/TIFS.2011.2129512. DOI: https://doi.org/10.1109/TIFS.2011.2129512
Arevalo-Ancona, R. E., Cedillo-Hernandez, M., & Garcia-Ugalde, F. J. (2024). Robust image tampering detection and ownership authentication using zero-watermarking and Siamese neural networks. International Journal of Advanced Computer Science and Applications, 15(10). https://doi.org/10.14569/IJACSA.2024.0151046 DOI: https://doi.org/10.14569/IJACSA.2024.0151046
Aydemir, G., Paynabar, K., & Acar, B. (2022). Robust feature learning for remaining useful life estimation using Siamese neural networks. In 2022 30th European signal processing conference (EUSIPCO) (pp. 1432-1436). IEEE.https://doi.org/10.23919/EUSIPCO55093.2022.9909630. DOI: https://doi.org/10.23919/EUSIPCO55093.2022.9909630
Chakraborty, U., Thilagavathy, D., Sharma, S. K., & Singh, A. K. (2024). Hybrid deep learning with AlexNet feature extraction and U-Net classification for early detection in leaf diseases. ICTACT Journal on Soft Computing, 14(3), 3255-3262. https://doi.org/10.21917/ijsc.2024.0457 DOI: https://doi.org/10.21917/ijsc.2024.0457
Darwish, M. M., Farhat, A. A., & El-Gindy, T. M. (2023). Convolutional neural network and 2D logistic-adjusted-Chebyshev based zero-watermarking of color images. Multimedia Tools and Applications, 83(10), 29969-29995. https://doi.org/10.1007/s11042-023-16649-3. DOI: https://doi.org/10.1007/s11042-023-16649-3
Dong, F., Li, J., Bhatti, U. A., Liu, J., Chen, Y. W., & Li, D. (2023a). Robust zero-watermarking algorithm for medical images based on improved NasNet-Mobile and DCT. Electronics, 12(16), 3444. https://doi.org/10.3390/electronics12163444. DOI: https://doi.org/10.3390/electronics12163444
Dong, S., Li, J., Bhatti, U. A., Ma, J., Dong, F., & Li, Y. (2023b). Robust zero-watermarking algorithm for medical images based on GFTT-KAZE and DCT. 26th ACIS International Winter Conference on Software Engineering, Artificial Intelligence, Net-working and Parallel Distributed Computing (SNPD-Winter). https://doi.org/10.1109SNPD-Winter57765.2023.10223753 DOI: https://doi.org/10.1109/SNPD-Winter57765.2023.10223753
Gong, C., Liu, J., Gong, M., Li, J., Bhatti, U. A., & Ma, J. (2022). Robust medical zero-watermarking algorithm based on Residual DenseNet. IET Biometrics, 11(2), 135-146. https://doi.org/10.1049/bme2.12100. DOI: https://doi.org/10.1049/bme2.12100
Han, B., Du, J., Jia, Y., & Zhu, H. (2021). Zero-watermarking algorithm for medical image based on VGG19 deep convolution neural network. Journal of Healthcare Engineering, (1), 5551520. https://doi.org/10.1155/2021/5551520. DOI: https://doi.org/10.1155/2021/5551520
Huang, T., Xu, J., Tu, S., & Han, B. (2023). Robust zero-watermarking scheme based on a depthwise overparameterized VGG network in healthcare information security. Biomedical Signal Processing and Control, 81, 1-10. https://doi.org/10.1016/j.bspc.2022.104478. DOI: https://doi.org/10.1016/j.bspc.2022.104478
Ikbal, F. & Gopikakamari, R. (2022). Performance analysis of SMRT-based color image watermarking in different color. Information Security Journal, 31(2), 157-167. https://doi.org/10.1080/19393555.2021.1873465. DOI: https://doi.org/10.1080/19393555.2021.1873465
Jing, W. (2020). November). Research on Digital Image Copying Watermarking Algorithm Based on Deep Learning. In 2020 International Conference on Robots & Intelligent System (ICRIS) (pp. 104-107). IEEE. https://doi.org/10.1109/ICRIS52159.2020.00034. DOI: https://doi.org/10.1109/ICRIS52159.2020.00034
Khafaga, D. S., Alhammad S. M., Magdi, A., Elkomy, O., Lashin, N. A., & Hosny, K. M. (2023). Securing transmitted color images using zero-watermarking and advanced encryption standard on Raspberry Pi. Computer Systems Science and Engineering, 4(2), 1967-1988. https://doi.org/10.32604/csse.2023.040345. DOI: https://doi.org/10.32604/csse.2023.040345
Lee, Y. H., Seo, Y. H., & Kim, D. W. (2019). Digital hologram watermarking by embedding Fresnel-diffracted watermark data. Optical Engineering, 58(6), 035102. https://doi.org/10.1117/1.OE.58.3.035102. DOI: https://doi.org/10.1117/1.OE.58.3.035102
Li, M., Liu, X., Wang, X., & Xiao, P. (2023a). Detecting building changes using multimodal Siamese multitask networks from very-high-resolution satellite images. IEEE Transactions on Geoscience and Remote Sensing, 61, 1-10. https://doi.org/10.1109/TGRS.2023.3290817. DOI: https://doi.org/10.1109/TGRS.2023.3290817
Li, Y., Li, J., Bhatti, U. A., Ma, J., Li, D., & Dong, F. (2023b). Robust zero-watermarking algorithm for medical images based on ORB and DCT. 26th ACIS International Winter Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD-Winter). https://doi.org/10.1109/SNPD-Winter57765.2023.10223992. DOI: https://doi.org/10.1109/SNPD-Winter57765.2023.10223992
Liu, Q. L., Yang, S. Q., Liu, J., Xiong, P. C., & Zhou, M. C. (2020). A discrete wavelet transform and singular value decomposition-based digital video watermark method. Applied Mathematical Modelling, 85, 273-293. https://doi.org/10.1016/j.apm.2020.04.015. DOI: https://doi.org/10.1016/j.apm.2020.04.015
Liu, Y. & Zhang, Z. (2021). Zero-watermarking algorithm based on DC component in DCT domain. 2021 International Conference on Electronic Information Engineering and Computer Science (EIECS). https://doi.org/10.1109/EIECS53707.2021.9588068.
Lin, T. Y., Maire, M., Belongie, S., Bourdev, L., Girshick, R., Hays, J., Perona, P., Ramanan, D., Zitnick, C. L., & Dollár, P. (2015). Microsoft COCO: Common objects in context. ArXiv. [En línea]. Disponible en: https://doi.org/10.48550/arXiv.1405.0312. Fecha de consulta: 15 de enero de 2025. DOI: https://doi.org/10.1007/978-3-319-10602-1_48
Shamia, D., Balasamy, K., & Suganyadevi, S. (2023). A secure framework for medical image by integrating watermarking and encryption through fuzzybased ROI selection. Journal of Intelligent & Fuzzy Systems, 44, 7449-7457. https://doi.org/10.3233JIFS-222618. DOI: https://doi.org/10.3233/JIFS-222618
Shang, C., Xue, Y., Liu, W. X., & Liu, Y. (2023). Visual image digital watermarking embedding algorithm combining 3D Boolean CNN and Arnold technology. IAENG International Journal of Computer Science, 50(4), 1221-1231.
Solorzano, C. & Tsai, D. M. (2023). Watermark detection in CMOS image sensors using cosine-convolutional semantic networks. IEEE Transactions on Semiconductor Manufacturing, 36(2), 279-290. https://doi.org/10.1109/TSM.2023.3245606. DOI: https://doi.org/10.1109/TSM.2023.3245606
Vizváry, L., Sopiak, D., Oravec, M., & Bukovčiková, Z. (2019). Image quality detection using the Siamese convolutional neural network. In 2019 International Symposium ELMAR (pp. 109-112). https://doi.org/10.1109/ELMAR.2019.8918678. IEEE DOI: https://doi.org/10.1109/ELMAR.2019.8918678
Wiggers, K. L., Britto, A. S., Heutte, L., Koerich, A. L., & Oliveira, L. S. (2019). Image retrieval and pattern spotting using Siamese neural network. 2019 International Joint Conference on Neural Networks (IJCNN). [En línea]. Disponible en: https://doi.org/10.1109/IJCNN.2019.8852197. Fecha de consulta: 10 de febrero de 2025. DOI: https://doi.org/10.1109/IJCNN.2019.8852197
Wu, J. Y., Huang, W. L., Zuo, M. J., & Gong, L. H. (2020). Optical watermark scheme based on singular value decomposition ghost imaging and particle swarm optimization algorithm. Journal of Modern Optics, 19(7), 1-13. https://doi.org/10.1080/09500340.2020.1810346. DOI: https://doi.org/10.1080/09500340.2020.1810346
Xiang, R., Liu, G., Li, K., Liu, J., Zhang, Z., & Dang, M. (2023). A zero-watermark scheme for medical image protection based on style feature and ResNet. Elsevier, 86(A), 1-11. https://doi.org/10.1016/j.bspc.2023.105127. DOI: https://doi.org/10.1016/j.bspc.2023.105127
Ye, H., Huang, X., Zhu, H., & Cao, F. (2025). An enhanced network with parallel graph node diffusion and node similarity contrastive loss for hypers-pectral image classification. Digital Signal Processing, 158, 104965. https://doi.org/10.1016/j.dsp.2024.104965 DOI: https://doi.org/10.1016/j.dsp.2024.104965
Zhang, Z. (2021). Zero-watermarking algorithm based on DC component in DCT domain. 2021 International Conference on Electronic Information Engineering and Computer Science (EIECS). [En línea]. Disponible en: https://doi.org/10.1109/EIECS53707.2021.9588068. Fecha de consulta: 10 de enero de 2025. DOI: https://doi.org/10.1109/EIECS53707.2021.9588068
Zhong, X. & Shih, F. Y. (2019). A robust image watermarking system based on deep neural networks. ArXiv. [En línea]. Disponible en: https://doi.org/10.48550/arXiv.1908.11331. Fecha de consulta: 15 de enero de 2025.
Published
How to Cite
License
Copyright (c) 2025 Universidad Autónoma de Tamaulipas
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Accepted 2025-06-27
Published 2025-08-18
