一种基于点特征的异源SAR图像适配性分析方法

A Point-feature Method for Multi-source SAR Image Matching Suitability Analysis

适配性分析是合成孔径雷达(Synthetic aperture radar,SAR)图像匹配的重要研究内容,它研究图像是否适合选作基准图的问题.本文研究基于点特征配准方法的异源SAR图像适配性分析,以特征点的数目及稳健程度来度量SAR图像适配性,提出基于点特征的异源图像适配性评价算法,并构建出一种评价特征点稳定性的准则,以指导选取合适的退化模型.实验结果表明本文所提出的点特征适配性评价机制能有效地度量指定SAR图像的适配性,从而近似得出保障图源的适配性排序结果.

Matching suitability is an important factor for synthetic aperture radar （SAlt） imageregistration. It focuses on to what extent the provided images are suitable as the reference images. This paper proposes an approach to measure and judge the SAR image matching suitability based on point-feature registration methods. The approach can estimate the SAR image matching suitability by using the number and stability of feature points. Significantly, a suitability analysis algorithm is presented for multi-source SAR image matching based on point features. Furthermore, a steady-point criterion is proposed to facilitate selection on the degradation models in numerical examples. The results show that the proposed method in this paper can effectively measure the matching suitability of the given SAR images and provide the appropriate order of these images under the point-feature criterion.