A medical image registration techniques based on improved artificial immune algorithm

Medical image registration is a process to match and superimpose two or more images of the same scene taken from different times, different imaging devices, or different perspectives. The drawback of traditional image registration algorithm is a large number of operations and slow convergence speed. In this paper, in order to improve the convergence speed on medical image registration algorithms, the initialization groups was optimized with an improved artificial immune algorithm and in order to make the algorithm more stable, the similarity of the standard was tested with the normalized measure. Those would improve the ability of global optimization solution. Furthermore the feasibility of the improved algorithm was verified by simulation experiments.