Research Article
An image encryption algorithm based on neural network
@INPROCEEDINGS{10.4108/eai.6-6-2021.2307843, author={Xuejun Li and Jiawu Yu and Huizhen Yan}, title={An image encryption algorithm based on neural network}, proceedings={Proceedings of the 8th EAI International Conference on Green Energy and Networking, GreeNets 2021, June 6-7, 2021, Dalian, People’s Republic of China}, publisher={EAI}, proceedings_a={GREENETS}, year={2021}, month={8}, keywords={laser chaotic system neural network arnold transform image encryption}, doi={10.4108/eai.6-6-2021.2307843} }- Xuejun Li
Jiawu Yu
Huizhen Yan
Year: 2021
An image encryption algorithm based on neural network
GREENETS
EAI
DOI: 10.4108/eai.6-6-2021.2307843
Abstract
Chaotic encryption provides a practical method for the confidentiality protection of today's digital images. Aiming at the problem of the secure transmission of image information in the network, an image encryption algorithm based on laser chaotic system and neural network is designed in this work. The phase diagram, Lyapunov exponential spectrum and bifurcation diagram analysis confirmed the chaotic characteristics of the single-mode laser power system. On this basis, the plaintext image is first compressed by the BP neural network, and then the compressed image is chaotically generated and diffused using the chaotic sequence obtained by the single-mode laser chaotic system to obtain the ciphertext image, and finally the histogram, Adjacent pixel correlation, anti-differential attack, information entropy and other aspects to test the security of the proposed encryption scheme. Simulation results show that the image encryption algorithm proposed in this work can not only save the channel bandwidth used in ciphertext transmission, but also have good encryption effect and high security performance.
