Image segment is one of the most fundamental and difficult problems in image analysis and computer vision. The image thresholding problem is treated as an important issue in image processing, and it can not only reduce the image data, but also lay a good foundation for succedent target recognition and image understanding. Character of global thresholding segmentation and local thresholding was analyzed in image segmentation. A new thresholding statistic iterative arithmetic is presented to overcome the direct worth method in thresholding, aiming at some lighting asymmetry and the abrupt a blemish for, or bigger arithmetic figure in ratio in a variety in gray of background image. Statistics iterative thresholding segmentation, based on image gray histogram and Gauss statistics distributing, obtain the theory expression of statistics iterative method and the best worth thresholding method and steps. Aviation image was thresholding segmentation using statistic iterative arithmetic, histogram technique and adaptive method respectively. Compared three thresholding results, it shows that statistic iterative method greatly improved the anti-noise capability of image segmentation and had the good result to the image of the worth and not easy to segment in full value thresholding method.