An image compression-encryption algorithm based on BP neural network optimized with fireworks algorithm

Area of Science:

• Computer Science
• Information Security
• Artificial Intelligence

Background:

• Increasing image data volume poses storage and transmission challenges.
• Existing algorithms struggle to balance compression efficiency with robust security.
• Need for advanced methods in secure image compression.

Purpose of the Study:

• To propose a novel algorithm for image compression and encryption.
• To enhance compression efficiency and reconstruction quality using an optimized BP neural network.
• To improve security through chaotic systems and advanced scrambling/diffusion techniques.

Main Methods:

• Optimized Backpropagation (BP) neural network using Fireworks Algorithm (FWA) for compression.
• Variable-parameter chaotic system (Logistic-Tent and Chebyshev maps) for pseudo-random sequence generation.
• Multi-level global-local scrambling and Gray Code-based mutation-diffusion for encryption.

Main Results:

• The optimized BP network improved compression efficiency and reconstruction quality.
• The chaotic system and scrambling mechanisms provided robust security.
• The proposed scheme outperformed standard BP networks in high-fidelity image recovery.

Conclusions:

• The novel algorithm effectively balances image compression and encryption.
• The integration of FWA-optimized BP networks and chaotic systems offers superior performance.
• The method demonstrates strong robustness and attack resistance for secure image handling.