An improved chaos-based image encryption scheme

Chaos theory provides a new approach to image encryption technology. The key stream generator is the key design issue of an image encryption system, it directly determines the security and efficiency of the system. This paper proposes an improved chaos-based key stream generator to enlarge the key space, extend the period and improve the linear complexity of the key stream under precision restricted condition so as to enhance the security of a chaos-based image encryption system. The generator is constructed by three Logistic maps and a nonlinear transform. The balance and correlation properties of the generated sequence are analyzed. The sequence is proved to be a binary Bernoulli stochastic sequence and the distribution of the differences between the amounts of 0 and I is analyzed. The side lobes of auto and cross correlation are proved to obey normal distribution N(0, 1/N). The experimental results indicate that the scheme has advantages of long period and strong anti various attack ability over conventional chaos-based encryption system.