基于改进的变分GAC模型矢量图像分割

Vector-valued images segmentation based on improved variational GAC model

在分析矢量图像颜色信息和动态曲线演化规律的基础上,将归一化传导率的非线性热方程约束项引入变分侧地活动轮廓矢量模型中,使水平集函数不用重新初始化即可快速稳定地保持符号距离函数的特性.改进算法减少了迭代次数和运行时间,改进了图像二维梯度和散度算子传统离散化方式,使梯度和散度算子保持空间旋转不变性,提高了分割算法的鲁棒性.实验表明该方法是有效的,对弱边缘具有较好的辨别能力.

An improved restriction item is introduced into variational GAC vector-valued model on the basis of analysis on color information and evolution characteristics of active contour curve. The proposed restriction item, which is a nonlinear heat equation with normalized diffusion rate, is added to the level set function to maintain the signed distance function properties fast and stably, and therefore the costly re-initialization procedure is completely eliminated. The algorithm reduces the number of iterations and run time. In addition, more efficient discretization method with spatial rotation-invariance gradient and divergence operator are proposed as numerical implementation scheme to improve strong robustness. The experiment results show that the proposed algorithm is effective and has the ability to distinguish the fuzzy edges.