基于Otsu与海域统计特性的SAR图像海陆分割算法
DOI:
作者:
作者单位:

作者简介:

通讯作者:

基金项目:


Se-Land Segmentation Algorithm of SAR Image Based on Otsu Method and Statistical Characteristic of Sea Area
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
    摘要:

    合成孔径雷达图像海陆分割在海面目标检 测、海岸线提取等海洋应用方面具有重要的意义。针对合成孔径雷达图像的特点,本文了提 出粗阈值与精确阈值相结合的海陆分割算法,并提供了一套完整的海陆分割方案。首先 利用改进的Otsu法计算粗阈值,获得大致海域范围;然后根据海域统计特性近似高斯分布的 特点,在分析了高斯分布统计特性的基础上,结合粗阈值分割结果,提出了计算出精确的分 割阈值的方法,进而获得准确的海陆分割图像;最后通过去除杂散点进一步完善分割结果。 实验结果表明,本文提出的方法具有较高的处理精度和较快的处理速度。

    Abstract:

    The s ea-land segmentation of SAR image is important in the marine applicatio n s, such as target detection, coastline extraction, etc. According to the charact eristics of SAR image, this paper puts forward a threshold segmentation algorith m combining rough threshold with accurate threshold, and provides a complete set of sea land segmentation scheme. Firstly, Otsu method is used to calculate a ro u gh threshold, and then the rough sea area can be obtained. Secondly, according to the feature that statistical characteristic of sea area similar to Gauss dist ribution, an accurate threshold can be calculated on the basis of rough threshol d segmentation, then the accurate sea land segmentation image can be obtained. F inally, [JP2]the segmentation result can be improved by the operation of removing spe ckle. The experiment results show that the proposed algorithm has high processin g precision and quick processing speed.[JP]

    参考文献
    相似文献
    引证文献
引用本文

陈祥,孙俊,尹奎英,于俊朋.基于Otsu与海域统计特性的SAR图像海陆分割算法[J].数据采集与处理,2014,29(4):603-608

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2014-09-02