华东师范大学学报(自然科学版) ›› 2006, Vol. 2006 ›› Issue (3): 66-70,1.

• 计算机科学 • 上一篇    下一篇

基于AOS格式的扩展Chan-Vese模型及多水平集图像分割方法

徐 盛, 李志斌   

  1. 华东师范大学 计算机系, 上海 200062
  • 收稿日期:2005-06-16 修回日期:2005-09-21 出版日期:2006-05-25 发布日期:2006-05-25
  • 通讯作者: 徐盛

Level Set Approach for Image Segmentation Based on Extended Chan-Vese Model and AOS Scheme(Chinese)

XU Sheng, LI Zhi-bin   

  1. Department of Computer Science, East China Normal University, Shanghai 200062, China
  • Received:2005-06-16 Revised:2005-09-21 Online:2006-05-25 Published:2006-05-25
  • Contact: XU Sheng

摘要： 对Chan-Vese提出的基于Mumford-Shah模型的多水平集图像分割算法作了改进.首先，使用AOS算法改进了原模型的差分格式，使得差分格式无条件稳定.其次，在水平集的构造中应用了快速推进法的改进算法，减少了水平集初始化时计算的点数和重复次数.

关键词: 图像分割, 扩展的C-V模型, 水平集方法, AOS算法, 符号距离函数, 图像分割, 扩展的C-V模型, 水平集方法, AOS算法, 符号距离函数

Abstract: A new level set approach for image segmentation was proposed based on the piecewise-constant Mumford-Shah model developed by Chan and Vese, The approach used AOS scheme, which is unconditional stable, to discrete the level set function. Besides, extended fast marching method for constructing the signed distance function was introduced to reduce the amount of computation significantly.

Key words: extended Chan-Vese model, level set method, AOS scheme, signed distant function, image segmentation, extended Chan-Vese model, level set method, AOS scheme, signed distant function

中图分类号: