常值初始化的自适应活动轮廓模型
Adaptive active contour starting with a constant initialization

李纯明提出的水平集方法(LI模型)很好地解决了测地活动轮廓模型(GAC)的重新初始化问题，但该模型对边缘信息较弱或者灰度不均匀的图像分割结果仍然不理想. 针对这些问题，提出常值初始化的自适应活动轮廓模型，该模型中自适应力的系数包含了图像的灰度信息，从而提高了活动轮廓在演化过程中对模糊边界的识别能力； 同时，重新定义的外部能量，避免过分割现象. 实验结果验证了模型的有效性.
Level set method，proposed by Li Chunming (LI model)，solve the problem of the geodesic active contour model (GAC) which need to reinitialize，but the result of this model with weak edge or uneven gray-level image segmentation is still not ideal. To overcome these problems，a new adaptive active contour model with constant initialization is proposed in the paper. The coefficient of the proposed model contains the mean gray value of the image，so that improve the active contour in the process of evolution of fuzzy boundary identification ability. At the same time，to redefine the external energy to avoid over-segmentation phenomenon. The experimental results show that the effectiveness of the proposed model