东北大学学报(自然科学版) ›› 2013, Vol. 34 ›› Issue (5): 632-636.DOI: -

• 信息与控制 • 上一篇    下一篇

一种新的几何活动轮廓模型在图像分割中的应用

张萍,高立群,薛哈乐   

  1. (东北大学信息科学与工程学院,辽宁沈阳110819)
  • 收稿日期:2012-11-15 修回日期:2012-11-15 出版日期:2013-05-15 发布日期:2013-07-09
  • 通讯作者: 张萍
  • 作者简介:张萍(1972-),女,辽宁沈阳人,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).

A Novel Geometry Active Contour Applied to Image Segmentation

ZHANF Ping,GAO Liqun,XUE Hale   

  1. School of Information Science & Engineering,Northeastern University,Shenyang 110819,China.
  • Received:2012-11-15 Revised:2012-11-15 Online:2013-05-15 Published:2013-07-09
  • Contact: ZHANG Ping
  • About author:-
  • Supported by:
    -

摘要: 提出了改进的LBF模型(ILBF)及其图像分割算法.利用两种不同尺度参数的LBF模型分别描述局部和全局信息,并构造了新的能量函数.将局部熵引入到ILBF模型中,同时给出自动求取能量函数中权重参数ω的有效方法,用尺度参数σ较大的LBF模型替代LGIF模型中的C-V模型,较大σ值的LBF模型不仅具有全局特性而且具有局部特性;将进行数据处理后的局部熵引入到LGIF模型中,进而自动求取权重参数ω,克服了LGIF模型权重参数值的选取全程都需要人工参与的缺点.为了有利于计算机的自动求解和避免过多无用的循环迭代,提出了一种新的终止准则.

关键词: 图像分割, 几何活动轮廓, LBF模型, ILBF模型, 局部熵

Abstract: An improved LBF (ILBF) model which is applied to image segmentation was proposed, which constructed a new energy function. The local and global information were described by LBF model with different scale parameters. And local entropy notion was applied to ILBF model and weight parameter ω in energy function could also be got automatically. The LBF model with a larger scale parameter had both global and local characteristics, and it was used instead of CV model in LGIF model. The ILBF model overcame the shortcoming of LGIF model in which the weight parameter was calculated by artificial. In order to avoid too much useless cyclic iterations, a new stop criterion was presented too.

Key words: image segmentation, geometry active contour, LBF model, ILBF model, local entropy

中图分类号: