Medical image registration is an important subject in medical image processing and analysis,in which pulmonary image registration is an instance of non-rigid registration.In practice,large deformation occurs in lung during respiratory movement or in operation.Pulmonary image registration is an active and challenging research topic in the field of image registration,with wide applications and perspectives.In this paper,the process of pulmonary image registration is introduced,and the pulmonary image registration is divided into five modules:the deformation model,the similarity measurement,the regularization method,the optimization strategy and the evaluation method.Moreover,the present situation and methods applied to each module are summarized.By reviewing the research status and the key technologies,this study presents the difficulties and future research directions in lung image registration.