The usage of genetic algorithm in image registration has some shortcomings,such as slow convergence and premature,which will result in artifacts.To overcome these problems,an improved adaptive genetic algorithm(IAGA)was proposed.The algorithm regard mutual information as similarity measure,improve the setting of genetic parameter,adjust adaptively the probabilities of crossover and mutation during the evolutionary process.The convergence rate was accelerated and premature was avoided.Experimental results indicate that improved genetic algorithm can achieve better feasibility and accuracy.