Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (8): 242-244.DOI: 10.3778/j.issn.1002-8331.2010.08.070

• 工程与应用 • Previous Articles     Next Articles

Coronary angiogram image segmentation using genetic algorithm and fuzzy connectedness

SHI Li,YAN Can   

  1. School of Electrical Engineering,Zhengzhou University,Zhengzhou 450001,China
  • Received:2008-09-18 Revised:2008-12-16 Online:2010-03-11 Published:2010-03-11
  • Contact: SHI Li


师 黎,闫 灿   

  1. 郑州大学 电气工程学院,郑州 450001
  • 通讯作者: 师 黎

Abstract: Considering the advantages of fuzzy connectedness and the characteristics of coronary images,fuzzy connectedness is used for coronary image segmentation.Genetic algorithm is combined with fuzzy connectedness to solve the problem which the implementation of fuzzy connectedness is difficult.First,the specific point is applied as a seed.Then the fuzzy connectedness between any point and the seed is obtained using genetic algorithm according to the given affinity formula.At last,the threshold is set,and the points which fuzzy connectedness are bigger than threshold are shown as the points having original values,the others are set as background.The experiment results show that coronary artery can be extracted efficiently.

Key words: genetic algorithm, fuzzy connectedness, image segmentation, coronary angiogram

摘要: 鉴于模糊连接法分割连通物体的优势及冠脉造影(CAG)图像中血管连通的特点,用模糊连接法来分割CAG图像。模糊连接法在实现上存在困难,提出将遗传算法与模糊连接相结合来解决这一问题。首先指定目标物体上一点作为种子点。然后根据给定的模糊亲和度公式,用遗传算法求出任意点到种子点的模糊连接度。最后设定阈值,将模糊连接度高于阈值的点按原值显示,其他点置为背景,即可得到分割结果。实验结果表明,该方法能快速准确将血管提取出来,为二维CAG图像的自动诊断及可视化提供了依据。

关键词: 遗传算法, 模糊连接度, 图像分割, 冠状动脉造影

CLC Number: