Image Encryption via DDPM Skipping-Step Diffusion Based on a Six-Dimensional Hyperchaotic System with Feedback...

Area of Science:

• Cryptography and Information Security
• Applied Mathematics
• Computer Science

Background:

• Traditional image encryption methods face challenges in balancing security and efficiency.
• The need for robust data protection in cloud-IoT environments is growing.
• Hyperchaotic systems offer complex dynamics, while diffusion models excel at data transformation.

Purpose of the Study:

• To develop a novel, highly secure image encryption algorithm.
• To leverage the strengths of hyperchaotic systems and diffusion models.
• To enhance plaintext sensitivity and ciphertext randomness for improved security.

Main Methods:

• Integration of a six-dimensional hyperchaotic system with denoising diffusion probabilistic models (DDPMs).
• Implementation of dual-level confusion-diffusion operations at pixel and bit levels.
• Utilizing the DDPM's Markov chain-based skipping-step diffusion mechanism.

Main Results:

• Achieved superior key space expansion and histogram uniformity.
• Demonstrated significant reduction in adjacent pixel correlation and optimized information entropy.
• Exhibited strong resilience against brute-force, statistical, and chosen-plaintext attacks.

Conclusions:

• The proposed algorithm offers a highly secure and practical solution for image data protection.
• The synergistic combination of hyperchaotic systems and DDPMs enhances encryption robustness.
• The method is well-suited for securing sensitive data in cloud-IoT applications.