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 ›› 2007, Vol. 33 ›› Issue (04): 176-178. doi: 10.3969/j.issn.1000-3428.2007.04.061

• 人工智能及识别技术 • 上一篇    下一篇

基于轮廓特征的SAR图像自动配准

刘小军1,周 越1,凌建国1,2,沈红斌1,杨 杰 1   

  1. (1. 上海交通大学图像处理与模式识别研究所,上海 200240;2. 中国航天科工集团公司二院,北京 100854)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-02-20 发布日期:2007-02-20

Patch-based SAR Image Automatic Registration

LIU Xiaojun1, ZHOU Yue1, LING Jianguo1,2, SHEN Hongbin1, YANG Jie1   

  1. (1. Inst. of Image Processing and Pattern Recognition, Shanghai Jiaotong University, Shanghai 200240; 2. Second Academy, China Aerospace Science Industry Corp, Beijing 100854)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-02-20 Published:2007-02-20

摘要: 针对SAR图像的自动配准问题,提出了一种基于轮廓特征的快速配准方法。在对SAR图像正确检测和分割的基础上提取图像的区域轮廓,并对其几何特征属性进行准确描述,利用配准点可以实现自动图像配准,仿真试验表明可以获得亚像素级配准结果。

关键词: 合成孔径雷达, 图像配准, 轮廓特征匹配, 特征提取

Abstract: To realize automatic registration of SAR image, an approach of fast match based on patch is proposed. Based on the correct region detection and segmentation, patches are extracted and their geometric features are accurately described, so image is registered by homologous points. Application of the proposed automatic registration model shows that it achieves sub-pixel level results.

Key words: SAR, Image registration, Patch-based match, Feature extraction