Image scrambling based on life-like cellular automata

This paper presents an image scrambling method based on the life-like cellular automata (CA). In the scrambling process, a life-like CA with an initial random configuration is set to run for several generations to obtain scrambling matrices. The restoration process is the reverse process of scrambling. In order to achieve a good diffusion property, we analyze how the scrambling effect is influenced by different CA initial configurations, and give the results of which life-like rules generate the best scrambling effect.