Abstract: The accurate segmentation of lung segmentation in CT image is the basis for the detection of pulmonary nodules, which is of great significance for diagnosis of clinical pulmonary diseases. Firstly, in this paper, the segmentation methods of lung segmentation based on CT image were reviewed. Then the main steps of the lung segmentation were detailed introduced, and discussed several typical algorithms of segmentation results, including the comparison of pulmonary parenchyma and pulmonary trachea. Finally, based on those methods, the paper integrated and improved several kinds of common segmentation method and algorithm, and proposes a practical and robust algorithm of lung segmentation.