欢迎访问中国科学院大学学报,今天是

中国科学院大学学报 ›› 2013, Vol. 30 ›› Issue (2): 238-243.DOI: 10.7523/j.issn.1002-1175.2013.02.015

• 信息与电子科学 • 上一篇    下一篇

基于模糊水平集的SAR图像分割方法

毛万峰1,2, 张红1, 张波1, 王超1   

  1. 1. 中国科学院对地观测与数字地球科学中心, 数字地球重点实验室, 北京 100094;
    2. 中国科学院研究生院, 北京 100049
  • 收稿日期:2012-03-09 修回日期:2012-04-19 发布日期:2013-03-15
  • 通讯作者: 张波
  • 基金资助:

    国家自然科学基金(40971198)资助

A new fuzzy level set method for SAR image segmentation

MAO Wan-Feng1,2, ZHANG Hong1, ZHANG Bo1, WANG Chao1   

  1. 1. Center for Earth Observation and Digital Earth Chinese Academy of Sciences, Beijing 100094, China;
    2. Graduate University, Chinese Academy of Sciences, Beijing 100049, China
  • Received:2012-03-09 Revised:2012-04-19 Published:2013-03-15

摘要:

提出一种SAR图像分割方法,即整合了模糊C均值聚类和基于区域水平集演化的分割方法.该方法通过模糊聚类的结果计算水平集演化的初始化条件及控制参数,从而克服了水平集演化依赖于初始化条件和控制参数且需要较多人工干预的缺陷,增强了方法的鲁棒性.模拟图像及真实SAR图像的实验表明,该方法在不需要人工干预的情况下,能够快速、准确地分割出感兴趣区域.

关键词: SAR, 图像分割, 模糊聚类, 水平集

Abstract:

We present a new method which integrates fuzzy c-means cluttering and region-based level set evolution for SAR image segmentation. Benefited by spatial fuzzy clustering, the initial level set segmentation approximates the component of interest. The controlling parameters are also estimated on the basis of the results of the spatial fuzzy clustering. The proposed method was evaluated on synthetic and real SAR images, and the results show that the new method is more robust, fast, and accurate in segmentation and does not need manual intervention.

Key words: SAR, image segmentation, FCM, level set

中图分类号: