Securing Images with Chaotic Map and Matrix Based Encryption

Resumen

This paper presents a novel grayscale image encryption scheme that combines
a high-dimensional chaotic map with a Suslin matrix-based diffusion process
to achieve robust security and high computational efficiency. In the proposed
method, a Hybrid Sinusoidal-Logistic-Chaotic (HSLC) map is employed to generate
a pseudo-random sequence used for pixel permutation, thereby effectively
disrupting spatial correlations in the plain image. Concurrently, a Suslin matrix
is constructed based on parameters derived from a shared secret key, which
is then used to perform block-wise diffusion on the permuted image. The diffusion
process operates on non-overlapping 4×4 pixel blocks, ensuring that even
minor changes in the plain image propagate widely in the cipher image. Security
analysis, including evaluations of Shannon entropy, correlation coefficients,
NPCR, UACI, and PSNR under noise attacks, demonstrates that the encrypted
images exhibit high randomness, strong differential resistance, and low pixel
correlation. Experimental results confirm that the proposed scheme is effective
in securing images for real-time multimedia applications.