FastChex Image Encryption Using Logistic-Sine Permutation and Dual XOR Diffusion

Abstract

Introduction: Chaos-based image encryption schemes often face challenges such as limited diffusion, high computational load due to block-based processing, and lack of adaptability, which hinder their use in real-time and low-resource environments. To address these issues, this paper presents ChaoxCrypt, a novel and efficient image encryption framework.

Methods: ChaoxCrypt employs a hybrid chaotic map (logistic–sine–tent) for pixel-level permutation, Gray code transformation for bit-level confusion, and a key-dependent adaptive S-box for substitution within the region of interest (ROI). In addition, AES encryption is applied using a dynamically generated key based on image content, increasing the randomness and variability of the encryption process.

Results: he proposed scheme demonstrates a Number of Pixel Change Rate (NPCR) above 99.9% and Unified Average Changing Intensity (UACI) between 20% and 33%, indicating strong diffusion and intensity variation. A high entropy value of 7.99 confirms its robustness against statistical attacks. Unlike traditional block-based approaches, ChaoxCrypt also achieves reduced memory usage and faster processing time.

Conclusions: With its integration of multiple encryption layers and adaptive strategies, ChaoxCrypt offers strong security, efficiency, and perfect reversibility. It is particularly suited for applications such as medical image protection, mobile communication, and video surveillance.