作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2012, Vol. 38 ›› Issue (12): 17-21. doi: 10.3969/j.issn.1000-3428.2012.12.005

• 专栏 • 上一篇    下一篇

医学图像分割中基于数据浓缩的谱聚类算法

丁 阳1,钱鹏江2?   

  1. (1. 无锡市第四人民医院放疗科,江苏 无锡 214062;2. 江南大学数字媒体学院,江苏 无锡 214122)
  • 收稿日期:2011-09-19 出版日期:2012-06-20 发布日期:2012-06-20
  • 作者简介:丁 阳(1977-),男,工程师,主研方向:图像处理,网络安全;钱鹏江,副教授、博士
  • 基金资助:

    国家自然科学基金资助项目(60903100, 60975027);江苏省自然科学基金资助项目(BK2009067)

Data Condensation Based Spectral Clustering Algorithm for Medical Image Segmentation

DING Yang 1, QIAN Peng-jiang 2   

  1. (1. Department of Radiation Oncology, Wuxi No.4 People’s Hospital, Wuxi 214062, China; 2. School of Digital Media, Jiangnan University, Wuxi 214122, China)
  • Received:2011-09-19 Online:2012-06-20 Published:2012-06-20

摘要:

基于传统Parzen窗密度估计函数的均值漂移谱聚类算法的时间复杂度不低于O(N2),不适合医学图像分割的实际需求。为此,通过压缩集密度估计和吸引盆均匀抽样两重数据浓缩策略以降低原MSSC的高时间开销问题,从而提出新的基于数据浓缩的谱聚类算法。实验结果表明,该算法能有效降低时间开销,较好地适应医学图像分割的要求。

关键词: 密度估计, 均值漂移, 谱聚类, 时间复杂度, 医学图像, 数据浓缩

Abstract:

The time complexity of the Parzen Window(PW) based Mean Shift Spectral Clustering(MSSC) algorithm is not less than O(N2), which means that it is impractical for medical image segmentation. In is paper, the problem of heavy time cost of original MSSC is solved by using two strategies of data condensation: reduced set density estimator and random sampling from every attraction basin, and the novel Data Condensation Based Spectral Clustering(DCBSC) algorithm is proposed. Compared with MSSC, the time cost of DCBSC is decreased effectively, and the practicability of DCBSC for medical image segmentation is improved accordingly.

Key words: density estimation, Mean Shift(MS), spectral clustering, time complexity, medical image, data condensation

中图分类号: