3种采样方法对DEM质量影响的定量分析

Quantitative Analysis on the Influence of Three Sampling Methods on DEM Quality

通过编程实现数字高程模型(Digital Elevation Model,DEM)数据采集的3种方法(随机采样、规则采样和成层随机采样)的算法,并对某一地区格网大小为30 cm×30 cm的栅格格式的高程数据进行采样,使用多种插值方法对采样数据进行插值,然后计算出偏差值。通过对偏差值的分析,发现这3种采样方法中随机采样较差,规则采样适合于对相对高程要求较高的应用,而成层随机采样适合于对高程总体偏差要求较高的应用。

This research was designed to evaluate different sampling methods of collecting Digital Elevation model （DEM） source data. A variety of methods including simple random sampling, systematic sampling, stratified sampling, cluster sampling, and progressive sampling had been used to collect DEM source data. However, there was little quantitative information on sampling methods of collecting DEM source data. In practice, sampling method was determined by experience. A 30 em x 30 cm DEM was used to test the effect of three common used sampling methods （ simple random sampling, systematic sampling, stratified sampling）. Every sampling result of each sampling methods was interpolated using three ways including Kriging, Spline, and Inverse Distance Weighted （IDW） in order to eliminate influence of different interpolating methods. By a combined analysis of experimental results, demonstrating simple random sampling was less accurate than the others ; systematic sampling was suitable for applications that require precise relative altitudes ; stratified sampling was suitable for applications that require precise average altitudes.