Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (13): 154-157.DOI: 10.3778/j.issn.1002-8331.2010.13.046

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

Accelerated improvement on isoperimetric algorithm

LIU Qi,XING Zhi-dong,DONG Jian-min   

  1. Department of Mathematics,Northwest University,Xi’an 710069,China
  • Received:2008-11-24 Revised:2009-02-02 Online:2010-05-01 Published:2010-05-01
  • Contact: LIU Qi

等周图像分割算法的一种加速改进

刘 琦,邢志栋,董建民   

  1. 西北大学 数学系,西安 710069
  • 通讯作者: 刘 琦

Abstract: Because of the drawbacks of the isoperimetric algorithm applied in image segmentation,such as not making full use of the state vector from solving linear equations and large amount of calculations produced by second iteration,this paper presents the accelerated improvement,the k+1 iterative method,after analysing the state vector and the isoperimetric ratio about images of multi-objective.Analyses and experiments show that the improved method is more efficient than the original segmentation algorithm in iterative numbers and time.

Key words: isoperimetric algorithm, isoperimetric ratio, image segmentation, state vector

摘要: 在等周算法应用于图像分割中,由于不能充分利用求解线性方程组所得到的状态向量,且对于多目标的图像进行处理时采用的二分迭代产生较大计算量的不足,通过对多目标图像的状态向量和等周率进行分析,提出一种加速改进k+1分迭代方式,经分析和实验表明改进后方法比原算法分割在计算次数和计算时间上效率更高。

关键词: 等周算法, 等周率, 图像分割, 状态向量

CLC Number: