This article presents a novel augmented image encryption algorithm tailored for securing satellite images, addressing the critical need for robust protection of sensitive geographic data. Implementing Shannon’s principles of confusion and diffusion, the method begins by augmenting multiple plain images into a single large image, followed by a three-stage encryption process. Initially, the augmented image is separated into its three color channels, which are transformed into one-dimensional (1D) bit-streams, split and altered using the Gauss Circle map, and restructured via Fredkin Gates to enhance unpredictability. Subsequently, the bit-streams are converted into 1D bytes and \(2 \times 2\) matrices, processed through three systems incorporating hyperchaos-induced keys and dynamic Hill Cipher matrices for additional confusion and diffusion. The final stage combines these encrypted streams into one image while preserving the integrity of color data. The proposed method achieves strong security metrics, including an average Number of Pixels Change Rate (NPCR) of \(99.6115\%\) , a Unified Average Changing Intensity (UACI) of \(31.71\%\) , and high entropy values (e.g., 7.9989) for encrypted images, ensuring robust resistance to differential and statistical attacks. The encryption demonstrates computational efficiency with an encryption time of 0.2817s for \(256\times 256\) images and maintains a low Peak Signal-to-Noise Ratio (PSNR) of 8.1 dB, reflecting effective data obfuscation. This multistage chaos-based approach, leveraging Fredkin logic gates and hyperchaos-induced keys, significantly enhances security, scalability, and efficiency, making it ideal for high-stakes satellite imagery applications where data integrity and confidentiality are paramount.
