摘要
针对基本蜻蜓算法因进化运动不稳定,从而引发收敛速度减慢等不足,提出衰变扰动方法和三分扰动方法,采用衰变扰动方法提升基本蜻蜓算法的寻优能力,采用三分扰动方法增强蜻蜓群体的富集性,进而提出一种衰变三分扰动蜻蜓算法。针对基本Powell算法依赖于初始点和易陷入局部极值等不足,采用衰变三分扰动蜻蜓算法对Powell进行优化,将其用于医学图像配准优化技术中,采用互信息作为评价标准,提出一种衰变三分扰动蜻蜓算法优化Powell的医学图像配准方法。通过仿真实验表明,该方法提高了医学图像配准的精度和鲁棒性。
The decay disturbance method and the tripartite disturbance method are proposed for the problem that the evolutionary motion of the basic dragonfly algorithm is instable,which leads to the slow convergence speed and other shortcomings.The decay disturbance method is used to improve the optimization ability of the basic dragonfly algorithm,the tripartite disturbance method is used to enhance the enrichment ability of the dragonfly population,and then the dragonfly algorithm with decay tripartite disturbance is proposed.As the basic Powell algorithm depends on the initial point and is easy to fall into the local extremum,the dragonfly algorithm with decay and tripartite disturbances is used to optimize Powell,and applied to the medical image registration optimization technology.A medical image registration method is proposed,which utilizes the dragonfly algorithm with decay and tripartite disturbances to optimize Powell,and for which the mutual information is taken as its evaluation criteria.The simulation experimental results show that this method can improve the accuracy and robustness of medical image registration.
作者
杜晓昕
张剑飞
王一萍
金梅
梁伟
DU Xiaoxin;ZHANG Jianfei;WANG Yiping;JIN Mei;LIANG Wei(College of Computer and Control Engineering,Qiqihar University,Qiqihar 161006,China)
出处
《现代电子技术》
北大核心
2020年第4期149-152,156,共5页
Modern Electronics Technique
基金
国家自然科学基金资助项目(61403222)
黑龙江省教育厅基本科研业务专项：齐齐哈尔大学科研项目资助(135109241
135109249)
关键词
衰变扰动
三分扰动
蜻蜓算法
Powell
互信息
医学图像配准
decay disturbance
tripartite disturbance
medical dragonfly algorithm
Powell
mutual information
medical image registration
作者简介
杜晓昕(1983-),女,硕士研究生,副教授,研究方向为计算智能与医学图像智能处理;张剑飞(1974-),男,博士研究生,教授,研究方向为机器学习与数据挖掘;王一萍(1971-),女,硕士研究生,副教授,研究方向为群智能计算。
