Enhancing image steganography through a novel multi-level chaotic encryption framework integrated with advanced deep neural networks

Tuieb, Munthir Bahir; Ahmed, Mohammed Hussein; Shibeeb, Ahmed Kareem; Abdulhamed, Mohamed Abdulrahman

doi:10.47974/JDMSC-2357

Open Access Research Article

Enhancing image steganography through a novel multi-level chaotic encryption framework integrated with advanced deep neural networks

^*Munthir Bahir TuiebCorresponding authormunthir_b_t87@uomustansiriyah.edu.iqDepartment of Computer Science College of Education Mustansiriyah UniversityBaghdad, Baghdad, IraqView full profile → , Mohammed Hussein Ahmedmohammedalbawi@uomustansiriyah.edu.iqDepartment of Computer Science College of Education Mustansiriyah UniversityBaghdad, IraqView full profile → , Ahmed Kareem Shibeebahmed.kareem@mtu.edu.iqDepartment of Computer Systems Technical Institute of Suwaira Middle Technical UniversityBaghdad, IraqView full profile → , Mohamed Abdulrahman Abdulhamedmohammed@uobasrah.edu.iqDepartment of Computer Science College of Computer Science and Information Technology University of BasrahBasrah, IraqView full profile →

* Corresponding author · click or hover a name for details

pp. 1947–1956Vol. 29Issue 5May 2026DOI: 10.47974/JDMSC-2357 Crossmark XML

Received:: 01 May 2025
Published Online:: 22 May 2026
Article type:: Research Article
Language:: EN
Article no.:: JDMSC-2357
Pages:: 1947–1956

Abstract

Undetectable capabilities, capacity, and security are the biggest challenges facing information hiding in images. This paper presents methods to address these challenges. It utilizes an advanced deep learning architecture combined with a chaotic, multi-level encryption system. Lorenz and Rossler maps are used as chaotic methods to encrypt the message, which is then segmented to increase its resistance to chaotic analysis. High-level features are extracted using ResNet-50 CNNs, and the image is classified into four regions based on entropy. This results in an optimal payload distribution (16x16 to 2x2). The method employs two techniques: Least Significant Bits Plus Plus (LSB++) for spatial modulation in textured regions and Discrete Cosine Transform (DCT) in smooth regions, which increases capacity and detectability. The results were obtained by evaluating two datasets, BOSSBase and COCO, with an average load of 4.2 bits per pixel, a signal-to-noise ratio (PSNR) higher than 54 dB, a structural similarity index (SSIM) higher than 0.97, and a modulation detection error rate of 89% using SRNet. The proposed method demonstrates significant improvements in modulation capacity and detection resistance compared to current methods.

Keywords

Steganography Chaotic encryption Advanced deep neural networks Least significant bit plus-plus Discrete cosine transform (DCT)

Subject Classifications

94A6068P30

References

[1] A. M. Alhomoud, “Image steganography in spatial domain: Current status, techniques, and trends,” Intelligent Automation & Soft Computing, vol. 27, no. 1 (2021).
[2] U. Pilania, M. Kumar, and V. Mittal, “Performance evaluation of LSB steganography based on multiple image formats,” Journal of Discrete Mathematical Sciences and Cryptography, vol. 26, no. 7, pp. 1911–1924 (2023).
[3] S. Wang, C. Wang, and C. Xu, “An image encryption algorithm based on a hidden attractor chaos system and the Knuth–Durstenfeld algorithm,” Opt. Lasers Eng., vol. 128, p. 105995 (2020).
[4] D. Sharma, C. Prabha, A. Goyal, M. Malik, and R. Kumar, “Enhanced security of EMI using an edge-based steganography with four-layered cryptography,” J. Discrete Math. Sci. Cryptogr, vol. 27, no. 2-B, pp. 775–784 (2024).
[5] X. Jin, X. Duan, H. Jin, and Y. Ma, “A Novel Hybrid Secure Image Encryption Based on the Shuffle Algorithm and the Hidden Attractor Chaos System,” Entropy, vol. 22, no. 6, p. 640 (2020).
[6] P. Wu, Y. Yang, and X. Li, “Stegnet: Mega image steganography capacity with deep convolutional network,” Future Internet, vol. 10, no. 6, p. 54 (2018).
[7] K. S. Kumar, K. B. Raja, and S. Pattnaik, “Hybrid domain in LSB steganography,” Int. J. Comput. Appl., vol. 19, no. 7, pp. 35–40 (2011).
[8] H. T. Elshoush, B. M. Al-Tayeb, and K. T. Obeid, “Enhanced Serpent algorithm using Lorenz 96 Chaos-based block key generation and parallel computing for RGB image encryption,” PeerJ Comput. Sci., vol. 7, p. 812 (2021).
[9] X. Wang and Y. Zhang, “An efficient method for the construction of block cipher with multi-chaotic systems,” Nonlinear Dyn., vol. 73, no. 1–2, pp. 53–61 (2013).
[10] M. Boroumand, M. Chen, and J. Fridrich, “Deep Residual Network for Steganalysis of Digital Images,” IEEE Transactions on Information Forensics and Security, vol. 14, no. 5, pp. 1181–1193 (2018).
[11] L. Li, W. Zhang, C. Qin, K. Chen, W. Zhou, and N. Yu, “Adversarial batch image steganography against CNN-based pooled steganalysis,” Signal Processing, vol. 181, p. 107920 (2021).
[12] D. Wang, G. Yang, J. Chen, and X. Ding, “GAN-based adaptive cost learning for enhanced image steganography security,” Expert Syst. Appl., vol. 249, p. 123471 (2024).
[13] S. M. Saber, M. B. Tuieb, K. K. Jabbar, M. H. Ahmed, and F. N. Abbas, “Utilizing variable hiding centers and dynamic block sizes to improve image steganography,” AIP Conf. Proc., vol. 3264, no. 1, p. 30020 (2025).
[14] A. K. Sahu, G. Swain, and E. S. Babu, “Digital image steganography using bit flipping,” Cybernetics and Information Technologies, vol. 18, no. 1, pp. 69–80 (2018).

Views: 69Downloads: 9Citations: 0