Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (34): 166-168.DOI: 10.3778/j.issn.1002-8331.2008.34.051

• 图形、图像、模式识别 • Previous Articles     Next Articles

C-V model using regularized level set function

YANG Yong1,PAN Wei-min1,XU Chun2   

  1. 1.Dept. of Computer,College of Math & Physics & Information Sciences,Xinjiang Normal University,Urumqi 830054,China
    2.College of Computer Science and Technology,Xinjiang University of Finance & Economics,Urumqi 830012,China
  • Received:2008-06-25 Revised:2008-09-16 Online:2008-12-01 Published:2008-12-01
  • Contact: YANG Yong

水平集函数规则化的C-V主动轮廓模型

杨 勇1,潘伟民1,徐 春2   

  1. 1.新疆师范大学 数理信息学院 计算机系,乌鲁木齐 830054
    2.新疆财经大学 计算机科学与工程学院,乌鲁木齐 830012
  • 通讯作者: 杨 勇

Abstract: C-V active contour model is proposed by Chan and Vese,which is implemented by the traditional level set method.In order to keep the evolution stable,the Euclidean length of the contour should be incorporated into regularize the level set function,and it also should be periodically re-initialized to be a signed distance function during the evolution,and this is expensive computational cost.A novel regularized method of the level set function is presented to keep its evolution stable.Moreover,re-initialization is unnecessary.Experimental results show the proposed method is robust and efficient.

摘要: Chan与Vese 提出的C-V主动轮廓模型采用传统的水平集方法实现,为了保证水平集函数演化的稳定性,需要加入轮廓的长度项来规则化水平集函数,且在演化的过程中要周期性重新初始化为符号距离函数,从而大大增加了计算量和实现的复杂度。提出一种新的规则化水平集函数的方法,不但可以保证水平集函数演化稳定,而且避免了重新初始化。实验结果表明:该方法稳健、快速。