Abstract: Aiming at the disadvantages of AES（Advanced Encryption Standard） encryption algorithm in the field of image encryption, an image encryption algorithm based on improved Henon hyperchaotic system combined with AES algorithm is proposed. Firstly, the algorithm uses the four-dimensional Henon hyperchaotic system to produce hyperchaotic sequences. By introducing the average value of the pixels of the plaintext image as the parameters to intercept the hyperchaotic sequence as the training sample of the BP neural network, and the nonlinear chaotic sequence obtained after the training is used as the goal key of the AES. Then, the target key is substituted into the AES encryption algorithm for two cycles of encryption to get the ciphertext. During each cycle of encryption, and the new S-boxes and round keys are generated from the chaotic sequence, greatly increasing the randomness of the key. The simulation results show that the proposed algorithm can well combine the advantages of the algorithms, which can achieve good encryption effect.

Key words: Henon hyperchaotic system, BP neural network, AES encryption algorithm, image encryption

摘要： 针对近年来AES（Advanced Encryption Standard）加密算法在图像加密领域应用中存在的一些缺点，提出了一种基于改进Henon超混沌系统与AES结合的图像加密算法。该算法首先利用四维Henon超混沌系统产生超混沌序列，通过引入明文图像像素的相关特性作为参数来截取超混沌序列作为BP神经网络的训练样本，训练后得到的非线性混沌序列作为AES加密算法的目标密钥。将目标密钥代入AES加密算法进行两轮循环加密得到密文，且每轮加密过程中都会由混沌序列产生新的S盒和轮密钥，大幅度提高了密钥的随机性。仿真实验结果表明，该算法能够很好结合三者的优点，安全性非常高。

关键词: Henon超混沌系统, BP神经网络, AES加密算法, 图像加密