节点文献
一种保持拓扑结构的隐式活动轮廓图像分割方法
A Topology Preserving Segmentation Method for Implicit Active Contour Models
【摘要】 提出一种保持拓扑结构的图像分割方法,通过应用一个限制拓扑结构的边缘检测函数替代经典的边缘检测函数,抑制了活动轮廓的拓扑结构变化.活动轮廓模型采用基于水平集方法的隐式结构,数值离散采用加性算子分裂(AOS)格式.结果表明,所提出的方法能有效地保持轮廓的拓扑结构,并且具有较高的计算效率.
【Abstract】 A novel topology preserving segmentation model was proposed.The proposed model prevents active contour models from geometrical or topological changes by using a topological constraint edge detector instead of the classical edge detector.The implicit scheme based on level set method is applied and an additive operator splitting(AOS) scheme is adopted.The result shows that the proposed method can effectively maintain the topology of the contours and cost less additional computing time.
【关键词】 图像分割;
水平集;
测地线活动轮廓;
保持拓扑结构;
限制拓扑结构;
加性算子分裂;
【Key words】 image segmentation; level set; geodesic active contour; topology preservation; topological constraint; additive operator splitting(AOS);
【Key words】 image segmentation; level set; geodesic active contour; topology preservation; topological constraint; additive operator splitting(AOS);
【基金】 吉林大学“985工程”项目基金
- 【文献出处】 吉林大学学报(理学版) ,Journal of Jilin University(Science Edition) , 编辑部邮箱 ,2009年05期
- 【分类号】TP391.41
- 【被引频次】4
- 【下载频次】157