Dynamics Analysis and Image Encryption of Memristor Neural Network with Sigmoid Activation Function

This paper studies a Hopfield neural network and proposes a memristive chaotic system (referred to as Memristive Sigmoid-HNN) with a Sigmoid function as the activation function. Under the influence of external electromagnetic radiation and applied current, the system simulates the impact of electromagnetic radiation by utilizing the fluctuation of magnetic flux on the neuron membrane. The dynamic behavior of the system was analyzed by changing the memristor parameters and applied current. The results indicate that the system exhibits abundant dynamic behaviors. The randomness of the system was tested, and the results show that the system has a high degree of randomness, making it suitable for image encryption. The system was used for image encryption, and the encryption results were analyzed, demonstrating that the system has good encryption performance.