中文摘要：

为了降低数字近景摄影测量中像面特征点匹配的复杂度,提高匹配的正确率,改善系统自动化水平,提出了一种基于空间光线聚集性的像面特征点匹配方法。该方法改变直接基于多二维像面的传统特征点匹配思路,在唯一的三维测量空间,根据像面特征点重建光线的聚集性,实现对应像面特征点的匹配,大大降低了匹配问题的复杂度。测量实验表明,该方法正确率高,可以满足大视场数字摄影测量系统要求。

英文摘要：

In order to reduce the complexity of feature point matching in digital close range photogrammetry, increase the matching ratio and improve the automation level of the system, a new target matching method based on spatial light clustering is proposed in this paper. In this method, the solution space is transformed from the traditional muhiple 2D spaces into the single 3D space. The feature points on the images are matched according to the aggregation of the reconstructed lights, and the complexity of matching problem is reduced effectively. The experimental results show that this new method has high matching ratio and meets the requirements of large scale digital photogrammetry.