Image registration is an important operation in many aerospace imaging applications, so that, in this paper a promising approach is proposed for satellite images registration. In this approach, we propose to use the Lorentzian norm as a fitness function with an artificial immune system optimization (AIS) technique to reduce the registration outlier and avoid trapping into local minima. To demonstrate the effectiveness of the proposed approach it is compared with the state-of-the-art Lukas-Kanade registration approach for many data sets.