计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (14): 195-197.

• 图形、图像、模式识别 • 上一篇    下一篇

基于遗传算法的形态学医学图像滤波

赵于前1,2,李桂源1,黄忠朝2   

  1. 1.中南大学 肿瘤研究所,长沙 410083
    2.中南大学 信息物理工程学院,长沙 410083
  • 收稿日期:2007-08-23 修回日期:2007-10-15 出版日期:2008-05-11 发布日期:2008-05-11
  • 通讯作者: 赵于前

Morphology medical image filtering based on genetic algorithm

ZHAO Yu-qian1,2,LI Gui-yuan1,HUANG Zhong-chao2   

  1. 1.Cancer Research Institute,Central South University,Changsha 410083,China
    2.School of Info-Physics and Geomatics Engineering,Central South University,Changsha 410083,China
  • Received:2007-08-23 Revised:2007-10-15 Online:2008-05-11 Published:2008-05-11
  • Contact: ZHAO Yu-qian

摘要: 滤波是医学图像处理的重要前期基础。针对形态学滤波的局限性,提出一种基于遗传算法的形态学滤波方法,该方法通过构造适当的适应度函数和对噪声样本图像的学习,利用遗传算法对结构元素的大小和形状以及中值滤波窗口进行自适应调整,找出最优的中值滤波窗和结构元素组合,从而获得最优的滤波效果。实验结果表明使用这一方法滤波后的医学图像信噪比得到了明显提高。

关键词: 遗传算法, 形态学, 医学图像, 信噪比

Abstract: Filtering is a very important preprocessing procedure for medical image.For the limitation of morphology filtering,the paper proposes genetic algorithm-based morphology filtering.First constructs appropriate fitness function and learned sample noise image,then adjusts adaptively the size and shape of structure elements and the median filtering window by genetic algorithm,the optimized combination of median filtering windows and structure elements is found,which can filter the medical image efficiently.The experimental result shows that the signal noise ratio of filtered medical image increases more by the proposed novel method than by morphology filtering and median filtering.

Key words: genetic algorithm, morphology, medical image, signal noise ratio