›› 2017, Vol. 2 ›› Issue (12): 49-52.

• 物理学 • 上一篇    下一篇

基于FCM聚类算法的人脑MRI图像分割

  

  1. (大理大学工程学院,云南大理671003)
  • 收稿日期:2017-07-14 出版日期:2017-12-15 发布日期:2017-12-15
  • 作者简介:谢勇,副教授,主要从事理论物理、生物医学工程学研究.
  • 基金资助:
    云南省教育厅科学研究基金资助项目(2015Y389)

Human Brain MRI Image Segmentation Based on FCM Clustering Algorithm

  1. (College of Engineering, Dali University, Dali, Yunnan 671003, China)
  • Received:2017-07-14 Online:2017-12-15 Published:2017-12-15

摘要: 磁共振成像(MRI)在脑功能的研究方面有着独特的优势,但脑组织的交织分布以及成像过程中磁场的不均匀性、部分容积效应以及噪声的影响,使得MRI图像的边界模糊不清。图像分割技术可以对医学图像中需要关注的区域进行分割处理,以辅助医生做出准确的诊断。在众多的医学图像分割技术中,模糊C-均值(FCM)聚类分割技术是目前广泛被使用的有效方法。讨论了基于MATLAB平台的医学图像FCM聚类的实现方法。

关键词: FCM聚类, 磁共振成像(MRI), 图像分割

Abstract: Magnetic resonance imaging(MRI)has a unique advantage in the study of brain functions, but the overlapping of brain tissues and the inhomogeneity of the magnetic fields in the process of imaging, the partial volume effects and noises make the MRI image boundary ambiguous. The image segmentation technology may segment and extract the needed area of medical image and assist the doctor in making an accurate diagnosis. The experimental results show that comparing with other medial segmentation technologies, the fuzzy C-mean(FCM)clustering technology can segment the human brain magnetic resonance imaging effectively. The method to realize the FCM clustering of medical images based on MATLAB is also discussed in the present research.

Key words: FCM clustering, Magnetic Resonance Imaging(MRI), image segmentation

中图分类号: