Image encryption based on Kronecker inner product over finite fields and adversarial neural network
Shengliang Zou, Weixia Xia, Gailin Zhu, Yaru Liang, Jianhua Wu · Journal of Physics Conference Series · 2021
Abstract An image encryption algorithm based on Kronecker inner product matrix over the finite field and adversarial neural network (ANN) is designed. The Kronecker inner product matrix transform and the ANN fulfil the tasks of confusion and diffusion simultaneously. In addition, the look-up table method is used to complete the addition and multiplication operations over GF(2 B ) finite fields, which can effectively improve the finite field computation speed while retaining its performance of non rounding errors. At the same time, relying on the secure hash function SHA-256 of the plain image to control the Logistic-Sine map greatly improves the diffusion and security of the encryption system. The simulation results verify that the proposed algorithm not only has high security and sufficient sensitivity, but also has a good resistance to various common attacks.