The week's pick
Random Articles
Reseach Article
GAN-based Adaptive Image Steganography
| International Journal of Computer Applications |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 187 - Number 6 |
| Year of Publication: 2025 |
| Authors: Dhanush Polisetty, Syed Wajahat Abbas Rizvi |
10.5120/ijca2025924951
|
Dhanush Polisetty, Syed Wajahat Abbas Rizvi . GAN-based Adaptive Image Steganography. International Journal of Computer Applications. 187, 6 ( May 2025), 45-50. DOI=10.5120/ijca2025924951
Abstract
Steganography through image hiding is one of the significant methods of secure communication. In the method of digital image hiding, the secret information is concealed without compromising its perceptibility. Classical steganographic techniques such as LSB-based approaches and DCT-based techniques face severe challenges regarding limited embedding capacity and susceptibility to steganalysis attacks. Here, a novel GAN-based steganography model that tries to find a balance between these two requirements and robustness against attacks has been proposed in this paper. We design GAN architecture for embedding secret data into the cover images such that these are left perceptually unchanged, and a discriminator that ensures the stego-images are indistinguishable from natural ones. The model is trained with a custom loss function that considers adversarial learning, perceptual quality, and embedding efficiency. Experimental assessment is performed on benchmark datasets, including COCO and Image Net, using the metrics of PSNR (Peak Signal-to-Noise Ratio), SSIM (Structural Similarity Index), and robustness. The results show that our GAN-based method surpasses traditional steganographic methods in terms of imperceptibility and resistance to steganalysis. Furthermore, the model remains robust against standard image transformations such as compression, noise addition, and cropping. This paper showcases the prospect of deep learning-driven steganography in the pursuit of improved data security and further proposes future improvements for real-world applications in secure communication and digital watermarking.
References
- I. Goodfellow, J. Pouget-Abadie, M. Mirza, et al., “Generative adversarial nets,” Advances in Neural Information Processing Systems, vol. 27, pp. 2672–2680, 2014.
- R. Zhang, P. Isola, A. A. Efros, E. Shechtman, and O. Wang, “The unreasonable effectiveness of deep features as a perceptual metric,” in Proc. IEEE CVPR, 2018, pp. 586–595.
- A. Dosovitskiy and T. Brox, “Generating images with perceptual similarity metrics based on deep networks,” in Proc. Advances in Neural Information Processing Systems, 2016, vol. 29.
- P. Isola, J. Zhu, T. Zhou, and A. A. Efros, “Image-to-image translation with conditional adversarial networks,” in Proc. IEEE CVPR, 2017, pp. 1125–1134.
- S. Baluja, “Hiding Images in Plain Sight: Deep Steganography,” Advances in Neural Information Processing Systems (NeurIPS), vol. 30, pp. 2063–2071, 2017.
- R. Zhang, Y. Li, and X. Luo, “Adversarial training for secure steganography,” IEEE Transactions on Information Forensics and Security, vol. 15, pp. 1502–1515, 2020.
- H. Huang, X. Zhao, R. Liu, et al., “Attention mechanism-based image steganography using GAN,” IEEE Access, vol. 9, pp. 115489–115499, 2021.
- Y. Zhang, H. Gao, and L. Liu, “Security evaluation of deep learning-based steganographic frameworks,” Multimedia Tools and Applications, vol. 81, pp. 3039–3061, 2022.
- X. Wu, Y. Zhang, and Y. Zhang, “A novel GAN-based image steganography method,” Neurocomputing, vol. 345, pp. 1–10, 2019.
- C. Ledig, L. Theis, F. Huszár, et al., “Photo-realistic single image super-resolution using a generative adversarial network,” in Proc. IEEE CVPR, 2017, pp. 4681–4690.
- Y. Geng, Y. Li, X. Luo, and S. Zhang, “Improving GAN training with multiple discriminators,” IEEE Transactions on Neural Networks and Learning Systems, vol. 30, no. 4, pp. 1305–1317, 2019.
- J. Hu, B. Wang, and Z. Shi, “Image steganography based on GAN with discriminator for steganalysis,” IEEE Access, vol. 9, pp. 12085–12093, 2021.
- J. Hayes and G. Danezis, “Generating steganographic images via adversarial training,” Advances in Neural Information Processing Systems, vol. 30, pp. 1954–1963, 2017.
- W. Tang, S. Tan, B. Li, and J. Huang, “Automatic steganographic distortion learning using a GAN,” IEEE Signal Processing Letters, vol. 24, no. 10, pp. 1547–1551, 2017.
- B. Wang, J. Hu, and J. Wang, “Steganography algorithm using a U-Net based generator,” Multimedia Tools and Applications, vol. 79, no. 19–20, pp. 13989–14010, 2020.
- H. Tancik, M. Mildenhall, T. Wang, et al., “Fourier features let networks learn high frequency functions in low dimensional domains,” Advances in Neural Information Processing Systems, vol. 33, pp. 7537–7547, 2020.
- X. Yang, H. Liu, Y. Cao, et al., “Robust image steganography with attention U-Net,” IEEE Transactions on Multimedia, vol. 23, pp. 1690–1703, 2021.
- N. Carlini, F. Tramer, E. Wallace, et al., “Wider and deeper, cheaper and faster: Tensor decomposition for steganography in neural networks,” arXiv preprint arXiv:1905.10985, 2019.
- Y. Liu, X. Li, and Y. Zhang, “Coverless image steganography using generative models,” Multimedia Tools and Applications, vol. 78, pp. 14353–14372, 2019.
- R. Tewari, M. Panda, and A. Abraham, “Deep learning-based steganography and steganalysis: A comprehensive review,” IEEE Access, vol. 10, pp. 1469–1495, 2022.
Index Terms
Keywords