计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (13): 145-146.DOI: 10.3778/j.issn.1002-8331.2010.13.043

• 图形、图像、模式识别 • 上一篇    下一篇

一种参数自动寻优的PCNN图像分割算法

卢桂馥1,2,王 勇1,窦易文1   

  1. 1.安徽工程科技学院 计算机科学与工程系,安徽 芜湖 241000
    2.南京理工大学 计算机学院,南京 210094
  • 收稿日期:2008-11-12 修回日期:2009-01-20 出版日期:2010-05-01 发布日期:2010-05-01
  • 通讯作者: 卢桂馥

Automated PCNN image segmentation method with optimal parameters

LU Gui-fu1,2,WANG Yong1,DOU Yi-wen1   

  1. 1.Department of Computer Science and Engineering,Anhui University of Technology and Science,Wuhu,Anhui 241000,China
    2.Institute of Computer Science,Nanjing University of Science and Technology,Nanjing 210094,China
  • Received:2008-11-12 Revised:2009-01-20 Online:2010-05-01 Published:2010-05-01
  • Contact: LU Gui-fu

摘要: 利用粒子群优化算法(Particle Swarm Optimization,PSO)具有对参数自动寻优的优势,将PSO和脉冲耦合神经网络(Pulse Couled Neural Network,PCNN)相结合,并以改进的最大类间方差准则函数为适应度函数,提出了一种能进行参数自动寻优的PCNN图像自动分割算法。实验仿真结果验证了该方法的有效性,即不仅可以正确地实现图像分割,而且PCNN的参数可以自动设置省去了人工实验的麻烦,同时分割速度也有所提高。

关键词: 脉冲耦合神经网络, 粒子群算法, 类间方差, 图像分割

Abstract: PSO algorithm has the ability to find the best parameters.So a new image segmentation method is proposed that bands PSO and PCNN and it’s objective function is an improved maximum between-cluster variance and it is used to segment the image automatically and successfully.The correctness and dependability of this method are verifiod by experiment results,that is to say,the quality of the segmentation method is much better and time-consuming is less and parameters-setting is automatical.

Key words: Pulse Couled Neural Network(PCNN), Particle Swarm Optimization(PSO) algorithm, between-cluster variance, image segmentation

中图分类号: