无人机影像像控离散点内插规则格网DEM数据精度评价

Accuracy assessment of interpolated regular grid DEM by discrete control points of UAV image

在利用离散点内插构建规则格网DEM的过程中,存在误差影响DEM的质量。对构建规则格网DEM的数据进行精度评价,为改善DEM质量提供一些参考依据。以样本点的选取和内插算法选择为切入点,对样本容量、样本点空间分布、格网间距和内插算法四个主要要素进行评价,实验表明,样本容量和样本点空间分布对内插精度的影响最为突出,要使DEM满足精度标准,需要足够大的样本容量和空间分布均匀的样本点;不同的DEM格网间距表示不同的比例尺,对于构建不同比例尺的DEM可以通过最佳数量的离散点构建;不同的内插算法有其适用范围和优缺点,内插时需要反复对比,选择适合的内插算法。

In the process of regular grid DEM construction by interpolation with discrete control points, some errors will reduce the quality of the DEM. Accuracy assessment of data in regular grid DEM construction provides some references to improve the quality of DEM. With the selection of sample points and interpolation algorithms as a starting point, the capacity and spatial distribution of sample points, the cell size of grid and interpolation algorithm are considered as the four main factors. The experimental results show that the capacity and spatial distribution of sample points affect interpolation accuracy obviously. For making the DEM accuracy meet standards, it requires sufficient capacity and spatial distribution of sample points. Different cell size of DEM means different scales, it may take optimal capacity of discrete points to construct DEM. Different interpolation algorithms have their own advantages and disadvantages, and application scope, it requires repetitive comparison to select the appropriate interpolation algorithm.