| 基于SIFT的多时相星载SAR图像特征点自动匹配 |
| |
| 引用本文： | 刘秀芳,尤红建.基于SIFT的多时相星载SAR图像特征点自动匹配[J].测绘科学,2009,34(1). |
| |
| 作者姓名： | 刘秀芳 尤红建 |
| |
| 作者单位： | 中国科学院电子学研究所,北京,100080 |
| |
| 基金项目： | 国家高技术研究发展计划(863计划) |
| |
| 摘 要： | SAR图像配准是实现多时相图像监测的前提,但是由于SAR图像的斑点噪声、成像机理的特殊性,使得采用常规的特征匹配方法很难实现SAR图像的自动配准。本文在介绍旋转尺度不变特征(SIFT)提取特征点原理的基础上,利用SIFT方法对两个时相星载SAR图像存在不同的旋转角和分辨率存在差异进行了特征点提取和自动匹配试验,图像特征点自动匹配的有效率达到70%以上,结果表明提取SIFT特征点进行多时相SAR图像自动匹配是基本可行的。
|
| 关 键 词： | SIFT 星载SAR 自动匹配 |
Matching feature points of multi-temporal space-borne SAR based on SIFT |
| |
| LIU Xiu-fang,YOU Hong-fian.Matching feature points of multi-temporal space-borne SAR based on SIFT[J].Science of Surveying and Mapping,2009,34(1). |
| |
| Authors: | LIU Xiu-fang YOU Hong-fian |
| |
| Abstract: | Registering multi-temporal SAR images automatically is the key preprocessing for monitoring,but it is difficult to register SAR images using traditional matching methods owing to the speckle noise and distinctive characteristic of SAR image.In this paper we introduce scale invariant feature transform(SIFT) algorithm to matching feature points in SAR image.The SIFT method is used to extract feature points in two temporal space-borne SAR images with different rotation angle and different resolution,the effective rate of successful matching reaches 70% and the results demonstrate that it is feasible to use SIFT feature points to register multi-temporal SAR image. |
| |
| Keywords: | SIFT |
| 本文献已被 万方数据 等数据库收录！ |
|
