几何约束和改进SIFT的SAR影像和光学影像自动配准方法

An Automatic Registration Algorithm for SAR and Optical Images Based on Geometry Constraint and Improved SIFT

提出一种基于几何约束和改进SIFT的SAR影像和光学影像自动配准方法。首先根据影像间的几何关系进行影像粗纠正,消除影像间旋转和分辨率差异;然后基于主方向改进的SIFT特征提取方法提取SIFT特征并利用其结构性信息引入结构相似性指数(SSIM)作为相似性测度获得初始匹配,经过视差空间和角度特征空间聚类优化得到稳定同名匹配;最后由随机抽样一致性算法(RANSAC)根据透视变换模型精化匹配结果获取变换模型参数。整个配准过程自动完成。本方法适用于差异较大的SAR影像与光学影像之间配准。

An automatic registration algorithm for SAR and optical images based on geometry constraint and improved SIFT isbroposed. Firstly a rough correction of the rotation and scale differences depending on the geometry constraint is applied. Then the SIFT features extracted by the dominant direction improved SIFT from two images are matched by SSIM as the similar measure according to the structure information of the SIFT feature. And then, parallax and angle restrictions are introduced to improve the matching performance by clustering analysis in the angle and parallax domains. Finally, the perspective transform parameters for the registration are obtained by RANSAC algorithm with removing the false matches simultaneously. The whole process is done automatically. The proposed algorithm is effective in the registration of SAR and optical images with large differences.