• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2013, Vol. 35 ›› Issue (6): 134-141.

• 论文 • 上一篇    下一篇

一种基于模板的港口舰船目标变化检测方法

随银岭,邹焕新,雷琳,李智勇   

  1. (国防科学技术大学电子科学与工程学院, 湖南 长沙 410073)
  • 收稿日期:2011-04-06 修回日期:2011-11-21 出版日期:2013-06-25 发布日期:2013-06-25

A novel change detection method of
harbor ship target based on template        

SUI Yinling,ZOU Huanxin,LEI Lin,LI Zhiyong   

  1. (School of Electronic Science and  Engineering,National University of Defense Technology,Changsha 410073,China)
  • Received:2011-04-06 Revised:2011-11-21 Online:2013-06-25 Published:2013-06-25

摘要:

系统地提出了一种以高分辨率光学图像为模板的变化检测思路和实现流程。首先,利用高分辨率光学图像制作二值化模板,将模板与待检测的两幅图像分别进行配准,在此基础上通过解算模板与序列图像间的空间变换关系来获取待检测区域;其次,将模板图与待检测图像进行加权融合获得分割海域图,并基于恒虚警检测算法提取变化的目标;最后,提取目标特征进行变化分析,得到变化信息。在实际的工程应用中,本方法得到了光学和SAR等不同源港口数据的有效验证。

关键词: 模板, 恒虚警检测, 变化检测, 变化分析

Abstract:

A novel change detection method based on template, which is made by high resolution optical image, was studied systematically in the paper. Firstly, high resolution optical images are used to make binary templates, the template and the two tobedetected images are matched separately, and then the region that waits for inspection is obtained by calculating the spatial transformation of the template and the image sequence. Secondly, the segmentation sea area images are obtained through weighted fusion of template and different temporal images,and then the CFAR algorithm is used to extract the changed targets. Finally, target features are extracted to perform change analysis. The efficiency of the method is proved by experimental results of different harbor images of SAR and optical in the practical application.

Key words: template;CFAR;change detection;change analysis