结合全局和双核局部拟合的活动轮廓分割模型

针对可缩放区域拟合（RSF）模型对初始轮廓敏感的缺点，提出了一种结合全局和局部图像信息的变分水平集活动轮廓模型。该模型设计了一个灰度域上的核函数，将其与RSF模型空域核的线性组合作为局部能量项，弥补了采样权值仅与空间距离有关的缺陷，提高了分割精度；构造了带有自适应全局指示函数的面积项，作为全局拟合力，提高了模型的收敛速度且避免陷入局部极小值；使用了高斯滤波方法，规则水平集函数，使其保持光滑，并避免了复杂的重新初始化过程。实验结果表明，该模型初始化灵活，对灰度不均匀图像有很好的分割效果

Active contour segmentation model of combining global and dual-core local fitting energy

As the Region-Scalable Fitting (RSF) model is sensitive to the location of initial curve, an active contour model combining global and local image information in a variational level set formulation was proposed in this paper. The local energy item was defined as a linear combination of the RSF model and our model by taking domain and range kernel functions into account, which made up for the defects that sampling weights were only related to spatial distance and improving the accuracy of segmentation. Also, it constructed an adaptive area term by introducing a global indicating function as a global intensity fitting force, which can speed up the convergence of the proposed model and can also avoid being trapped into local minima. In the numerical calculation, Gaussian filtering was utilized to regularize the level set function, which can ensure its smoothness and eliminate the requirement of re-initialization．The experimental results show that the proposed model allows flexible initialization, and it can achieve desirable results when being applied to images with inhomogeneous intensity.