东北大学学报(自然科学版) ›› 2013, Vol. 34 ›› Issue (2): 166-169.DOI: -

• 信息与控制 • 上一篇    下一篇

融合C-V和GVF的测地线活动轮廓模型

潘改1,高立群1,张萍2   

  1. (1.东北大学信息科学与工程学院,辽宁沈阳110819;2.鞍山师范学院物理科学与技术学院,辽宁鞍山114005)
  • 收稿日期:2012-08-31 修回日期:2012-08-31 出版日期:2013-02-15 发布日期:2013-04-04
  • 通讯作者: 潘改
  • 作者简介:潘改(1983-),女,安徽宿州人,东北大学博士研究生;高立群(1949-),男,辽宁沈阳人,东北大学教授,博士生导师.
  • 基金资助:
    国家自然科学基金资助项目(81000639).

Geodesic Active Contour Model Combined with CV and GVF

PAN Gai1, GAO Liqun1, ZHANG Ping2   

  1. 1. School of Information Science & Engineering, Northeastern University, Shenyang 110819, China; 2. Physics Science and Technology College, Anshan Normal University, Anshan 114005, China.
  • Received:2012-08-31 Revised:2012-08-31 Online:2013-02-15 Published:2013-04-04
  • Contact: PAN Gai
  • About author:-
  • Supported by:
    -

摘要: 对于有凹陷边界或弱边界的待分割目标,采用传统的测地线活动轮廓(GAC)模型无法进行准确的图像分割.为了解决这一问题,提出了一种融合C-V模型、GVF模型和GAC模型的图像分割算法.在该算法中,GAC模型的单位内法向量与GVF模型的梯度矢量流共同作用,促使轮廓曲线向目标的边界方向运动;而GAC模型单位内法向量与C-V模型的区域信息的力场共同作用,不仅促使轮廓曲线向目标的边界方向运动,而且使轮廓曲线稳定在目标的边界上.仿真实验证明了上述方法的有效性,同时还证明了该方法对轮廓曲线的初始位置具有较好的适应性.

关键词: 测地线活动轮廓, C-V模型, GVF模型, 梯度矢量流, 单位内法向量

Abstract: An image with concave edges or weak edges cannot be segmented precisely using the conventional geodesic active contour (GAC) method. So, a novel image segmentation algorithm was proposed by combining the CV and GVF models with the GAC model. In this algorithm, the unit inward normal of the GAC model was joined to the gradient vector flow of the GVF model, moving the contour curve towards the boundary of the object. Also, it was joined to the region information of the CV model, getting the curve not only to move to but also to stay at the boundary of the object. Simulation results show that the algorithm proposed is effective and robust to the initial location of the contour curve.

Key words: geodesic active contour, CV model, GVF model, gradient vector flow, unit inward normal

中图分类号: