上海交通大学学报

上海交通大学学报（自然版） ›› 2011, Vol. 45 ›› Issue (03): 301-0307.

• 无线电电子学、电信技术 •    下一篇

基于混沌弹性粒子群优化与基于分解的二维交叉熵阈值分割

  

  1. （1.南京航空航天大学 电子信息工程学院, 南京 210016； 2.光电控制技术重点实验室, 河南 洛阳 471009）
  • 收稿日期:2010-04-12 出版日期:2011-03-30 发布日期:2011-03-30
  • 基金资助:

    国家自然科学基金资助项目（60872065），航空科学基金资助项目(20105152026)

Twodimensional Cross Entropy Thresholding Based on Chaotic Resilient Particle Swarm Optimization or Decomposition

  1. (1. College of Electronic Information Engineering, Nanjing University of Aeronautics and Astronautics,  Nanjing  210016, China; 2. Key Laboratory on Electrooptic Control Technology, Luoyang 471009, Henan, China)
  • Received:2010-04-12 Online:2011-03-30 Published:2011-03-30

摘要： 为了提升二维交叉熵阈值分割法运行速度,提出了基于混沌弹性粒子群优化（CRPSO）和基于分解的2种二维交叉熵阈值分割算法.前者利用CRPSO算法寻找二维交叉熵法的最佳分割阈值,并采用递推方式避免迭代过程中适应度函数的重复计算,使运算速度大大提高;后者将二维交叉熵法的运算转换到2个一维空间上,计算复杂度由O(L2)进一步降为O(L).实验结果表明,2种算法能够在保证分割效果达到或优于现有二维交叉熵阈值分割法的前提下,运行时间大幅减少.

关键词: 图像分割, 阈值选取, 二维交叉熵, 混沌弹性粒子群优化

Abstract: A twodimensional cross entropy image thresholding method based on chaotic resilient particle swarm optimization (CRPSO) or decomposition was proposed. Firstly, chaotic resilient particle swarm optimization was used to find the optimal threshold of twodimensional cross entropy method. The recursive algorithm was adopted to avoid the repetitive computation of fitness function in iterative procedure. As a result, the computing speed was improved greatly. Then, the computation of twodimensional cross entropy method was converted into two onedimensional spaces, which made the computation complexity further reduce from O(L2) to O(L). The experimental results show that, the two methods proposed in this paper can greatly reduce the running time while the segmented result is as good as or better than the existing twodimensional cross entropy thresholding method.

Key words: image segmentation, threshold selection, twodimensional cross entropy, chaotic resilient particle swarm optimization（CRPSO）, decomposition

中图分类号: