针对斑点噪声对合成孔径(SAR)图像匹配算法的影响,提出了一种基于各向异性尺度空间的SAR图像匹配算法。首先,采用加性算子分裂算法解方案来构建各向异性尺度空间,在滤除斑点噪声的同时更好地保留图像细节;然后,在非线性尺度空间中提取特征点,并采用改进的SURF描述子描述特征,弱化斑点噪声对匹配的影响;最后,采用变换参数约束策略筛选匹配点对,提高匹配正确率。该方法既保持了同名点的精度还增加了同名点的数量,通过对不同极化、时相、波段以及不同视角下多种地物的匹配实验,验证了该方法的优越性。 Aiming at the effect of speckle noise on synthetic aperture (SAR) image matching algorithm, a SAR image matching algorithm based on anisotropic scale space is proposed. Firstly, the solution of the additive operator splitting algorithm is used to construct the anisotropic scale space, which preserves the image detail better while removing speckle noise. Secondly, the feature points are extracted in the non-linear scale space and the improved SURF Describe the characteristics of the sub-descriptors and weaken the influence of speckle noise on the matching. Finally, we use the transformation parameter constraint strategy to filter matching pairs and improve the matching accuracy. The method not only maintains the accuracy of the same name points but also increases the number of the same name points. The matching experiments of different kinds of features under different polarizations, time phases, wave bands and different perspectives verify the superiority of the method.