Abstract: In order to increase the stability of image segmentation and improve accuracy of image segmentation, this paper proposed an image segmentation method based on improved active contour model. Firstly, internal target contour is select as the initial point, and the initial point expand to boundary point to make more continuous and smooth edge, and secondly improved greedy algorithm is used to remove the shrinkage tendency of active contour, and prevent accumulation phenomenon of control points, finally, simulation experiment is used to analyze the effect of image segmentation. The results show that improved active contour model not only can achieve the accurate segmentation of image, but also has strong robustness to noise, and segmentation error is smaller than other image segmentation models, and has wide application prospects.