Skip to content

notashock/hsi-encryption

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

41 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

NeoCrypt-HSI: High-Speed 16-Bit Hyperspectral Image Encryption

NeoCrypt-HSI is a cutting-edge cryptographic framework designed specifically for the secure transmission and storage of 16-bit Hyperspectral Image (HSI) cubes. By integrating high-dimensional chaotic maps, DNA-inspired algebraic rules, and Quantum Key Distribution (QKD) seeding, this project provides "unconditional security" without compromising computational speed or data integrity.


🚀 Key Features

  • Native 16-Bit Support: Unlike standard 8-bit image encryption, this pipeline operates directly on 16-bit HSI data, preserving the high-fidelity spectral signatures essential for remote sensing and scientific analysis.
  • Lossless Data Integrity: Guarantees 100% bit-perfect restoration (MSE = 0, PSNR = ∞) between original and decrypted datasets.
  • Low-Spec Hardware Optimization: Uses Statistical Quantization (Rank Mapping) instead of heavy index-based sorting, allowing the algorithm to encrypt massive .mat files on consumer-grade hardware (e.g., Intel i3, 8GB RAM) without memory crashes.
  • Plaintext-Dependent Security: Leverages U-PEP (Universal Plaintext-Equivalent Perturbation) extraction to ensure that the encryption keys are intrinsically linked to the image content, providing total immunity against known-plaintext attacks.

🛠️ System Architecture

The encryption process follows a sophisticated three-stage volumetric pipeline:

1. Key Seeding & Chaos Generation

  • QKD Integration: The initial master key is seeded via simulated Quantum Key Distribution.
  • U-PEP Extraction: A SHA-512 hash of the image content is merged with the QKD key to generate highly sensitive chaotic parameters.
  • Chaotic Engine: A high-speed Numba-accelerated generator creates four distinct chaotic sequences (X, Y, Z, K).

2. Volumetric Scrambling

  • Spatial Confusion: Pixels are relocated using chaotic circular row and column shifts, breaking the visual structure of the HSI cube.
  • Spectral Scrambling: The "spectral signature" of each pixel is rotated across the Z-axis (bands), obfuscating the relationships between contiguous spectral frequencies.

3. DNA XOR Diffusion

  • 16-Bit Splitting: Pixel data is split into high and low bytes and encoded into DNA bases (A, T, C, G).
  • Dynamic Rule Selection: Eight distinct DNA algebraic XOR rules are applied pixel-by-pixel, driven by the chaotic Z-sequence, ensuring high sensitivity and massive entropy.

📊 Security & Performance

The methodology has been benchmarked against industry-standard metrics:

Metric Achieved Result Ideal Target
Information Entropy ~15.999x 16.0 (for 16-bit data)
NPCR > 99.6% > 99.6%
UACI ~ 33.4% ~ 33.4%
Correlation Coeff. ~ 0.000x 0.0 (No correlation)
Restoration MSE 0.0 0.0

💻 Technical Prerequisites

  • Python 3.8+
  • Core Libraries: numpy, scipy, hashlib, time, gc
  • Acceleration: numba (for JIT-compiled chaos cores)
  • Hardware Monitor: psutil (for dynamic RAM management)

📖 Usage

  1. Load Dataset: Import your HSI .mat file (e.g., Pavia University, Indian Pines).
  2. Run U-PEP & Chaos Cell: Extract keys and generate chaotic sequences. The system will automatically detect if your RAM is sufficient and apply Statistical Quantization if needed.
  3. Execute Encryption: Run the vectorized encryption cell to transform the image into ciphertext.
  4. Analyze Results: Use the Comprehensive Security Audit cell to verify entropy and avalanche effects.

🔮 Future Work

  • Exploration of QKD Integration Strategies: Investigate and evaluate various physical and logical frameworks for integrating Quantum Key Distribution into the existing 16-bit pipeline to establish an unconditionally secure key exchange layer.

License

This project is developed for research and academic purposes.

About

It is a project that is built to encrypt Hyperspectral images using Quantum Key Distribution

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors