摘要
在黄土高原选取代表不同地貌类型的延川、绥德、富县3个研究样区,以及1∶1万地形图所建立的不同水平分辨率数字高程模型(DEM)为研究对象,分别分析每一大样区不同水平分辨率DEM所提取坡度的面积误差变化规律.提出面积百分比加权总体误差和面积百分比平均总体误差的指标,在每一大样区随机选取20个试验样区,得到两项误差指标与水平分辨率的线性函数关系.根据线性函数关系,由已知误差求得所需要的水平分辨率范围,并在每一大样区的16个随机检验样区通过检验.线性函数关系模型的建立,可望为实际工作选取适宜的水平分辨率提供依据,对于在保证精度的条件下节约成本,减小数据量具有一定意义.文中最后提出,所指出的两项指标不适用于风沙-黄土过渡区域.
The author takes Yanchuan, Suide, and Fu County which are different in physiognomy in the Loess Plateau as the sample areas of the study and DEM with different horizontal resolutions is established according to the map scaled at 1 10 000 as the object of the research. For each sample area, the variation rules of the area errors of slopes derived from DEM with different horizontal resolutions are analyzed and the indexes of the total error of weighted area percentage and the total error of average area percentage are proposed for the first time. The linear relationship between the two indexes and different resolutions is obtained in 20 test areas. By adopting this model, the needed range of the horizontal resolution can be calculated. The model of the linear relationship has passed the test in 16 randomly selected test areas. The establishment of the model will provide basis for selecting proper horizontal resolutions in practice. It also has significance in decreasing the amount of data and cutting down the cost while ensuring the accuracy of slope information. The above two indexes are not appropriate for the transitional loess regions.
出处
《华侨大学学报(自然科学版)》
CAS
北大核心
2006年第3期321-324,共4页
Journal of Huaqiao University(Natural Science)
基金
国家自然科学基金资助项目(40571120
40301027)
福建省科技计划基金资助项目(2005Q009)