Parallel Image Encryption Technology Based on Cellular Automaton

The traditional encryption algorithms such as DES, AES are considered too slow for image encryption, which cannot meet the real-time requirements. Therefore, this paper proposes a parallel image encryption algorithm based on cellular automaton and chaotic map. In the scrambling stage, the keystream generated by 2D-LASM is used to scramble the rows and columns of the image in parallel. In the diffusion stage, the scrambled image is constructed into two identical square bit matrices as the initial state of the cellular automaton based on the balance rule, and the encrypted image is obtained by iterating cellular automaton multiple times. The sub key expansion algorithm adopts a weighted pixel value mechanism, which makes the generation of key stream depend on plain image, so as to resist the attack of chosen plaintext and known plaintext. The algorithm is coordinated by CPU / GPU. With the acceleration of GPU, the efficiency of the algorithm has been significantly improved.