Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (32): 237-240.

• 工程与应用 • Previous Articles     Next Articles

Medical image registration through Improved Particle Swarm Optimization

XIE Jing-quan1,2,XU Wen-bo1,SUN Jun1   

  1. 1.Institute of Information Technology,Southern Yangtze University,Wuxi,Jiangsu 214122,China
    2.Wuxi Institute of Technology,Wuxi,Jiangsu 214121,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-11 Published:2007-11-11
  • Contact: XIE Jing-quan

基于IPSO算法的医学图像配准

谢景权1,2,须文波1,孙 俊1   

  1. 1.江南大学 信息工程学院,江苏 无锡 214122
    2.无锡职业技术学院,江苏 无锡 214121
  • 通讯作者: 谢景权

Abstract: Medical image registration is the first step for image fusion and other imaging process.In this paper,The image edges are first detected by using Canny operator,then the contour feature points are extracted by K-means algorithm,and translation parameters are calculated by using a Improved Particle Swarm Optimization (IPSO) algorithm.Experiments show that this approach is efficient and can avoid local minimum.

Key words: medical image registration, canny operator, K-means clustering algorithm, Improved Particle Swarm Optimization(IPSO)

摘要: 医学图像配准是医学图像分析诊断的基础,也是图像融合等图像处理需要先行解决的问题。首先用Canny算子提取图像的边缘,再用K-Means聚类算法进行聚类分析提取轮廓特征点,然后提出了一种改进的粒子群优化(IPSO)算法来求解配准所需的空间变换参数。实验结果表明:改进PSO能够迅速地在全局范围内找到最优解,应用于多模态医学图像配准是可行的。

关键词: 医学图像配准, canny算子, K-Means聚类, IPSO