Chaotic-Map-Based Image Steganography Using Bernoulli and Logistic Functions

Malcoaccha-Díaz E.R.; Flores-Quispe R.; Velazco-Paredes Y. · International Journal of Computing and Digital Systems (2025) · DOI: 10.12785/ijcds/1571127293

Resumen

The exponential growth of digital information and increasing cybersecurity threats require robust protection mechanisms for sensitive visual data. While conventional steganography methods exist, many remain vulnerable to statistical attacks. This article proposes an enhanced hybrid security approach that combines the Least Significant Bit (LSB) steganography method with Bernoulli and Logistic chaotic maps, integrating complementary mathematical transformations to create a multi-layered security framework. A dual-layer protection mechanism—featuring chaotic pixel permutation encryption and randomized bit placement steganography—ensures its content remains unintelligible even if the hidden data is extracted without the corresponding chaotic keys. Extensive evaluation using advanced metrics—including Peak Signal-to-Noise Ratio (PSNR), Mean Squared Error (MSE), Normalized Absolute Error (NAE), Structural Similarity Index Measure (SSIM), Fourier Transform Analysis, and Chi-Square statistical testing—demonstrates superior performance compared to state-of-the-art methods. The experimental results yielded impressive metrics: average PSNR of 56.49 dB, SSIM values exceeding 0.9998, remarkably low MSE values (as low as 0.10), and Chi-square test p-values consistently below 0.01, establishing statistical imperceptibility between cover and stego-images. The proposed technique ensures hidden images remain imperceptible while preserving the structural integrity and quality of carrier images, offering an effective solution for safeguarding sensitive visual information in applications such as forensic evidence protection, secure communications, and digital watermarking.

Instituciones: Universidad Nacional de San Agustín de Arequipa