Related Experiment Video
Updated: Aug 9, 2025
10:00
Gradient Echo Quantum Memory in Warm Atomic Vapor
Published on: November 11, 2013
12.9K
Coupling Quantum Random Walks with Long- and Short-Term Memory for High Pixel Image Encryption Schemes.
Junqing Liang1, Zhaoyang Song1, Zhongwei Sun1
1School of Information and Control Engineering, Qingdao University of Technology, Qingdao 266033, China.
Entropy (Basel, Switzerland)
|February 25, 2023
Summary
This study introduces a novel image encryption scheme using quantum random walks and long short-term memory (LSTM) networks. This method enhances pseudorandom matrix generation for secure, high-pixel-density image encryption.
Area of Science:
- Cryptography
- Image Processing
- Artificial Intelligence
Background:
- Traditional encryption methods struggle with high pixel density images.
- Quantum random walk algorithms face efficiency challenges in generating large pseudorandom matrices.
Purpose of the Study:
- To propose an efficient and robust image encryption scheme for high pixel density images.
- To leverage Long Short-Term Memory (LSTM) networks to improve quantum random walk algorithm efficiency and pseudorandom matrix quality.
Main Methods:
- Integration of quantum random walk algorithm with LSTM networks for pseudorandom matrix generation.
- Training LSTM networks with image data to predict highly random output matrices.
- Generating an LSTM prediction matrix for image encryption based on pixel density.
Main Results:
- Achieved high information entropy (average 7.9992).
- Demonstrated strong pixel diffusion and confusion with high NPCR (average 99.6231%) and UACI (average 33.6029%).
- Exhibited low correlation (average 0.0032) between adjacent pixels.
Conclusions:
- The proposed scheme effectively encrypts high pixel density images.
- The integration of LSTM with quantum random walks enhances pseudorandom matrix generation for robust encryption.
- The encryption scheme shows resilience against noise and interference, suitable for real-world applications.
