DEM地形描述误差(Et)计算模型研究
作者简介:王 春(1975-),男,宁夏,博士,副教授,主要从事DEM数字地形分析和GIS集成应用技术研究。E-mail:wangchun93@126.com
收稿日期: 2014-02-10
要求修回日期: 2014-03-20
网络出版日期: 2014-09-04
基金资助
国家自然科学基金项目“DEM地形描述误差的区域差异性研究”( 41001301)
国家自然科学基金项目“多源多尺度DEM地表形态精度量化分析研究”(41201415)
安徽省自然科学基金项目“数字高程模型可用性量化分析研究”(1408085MD77)
A New Calculation Model of DEM Terrain Description Error
Received date: 2014-02-10
Request revised date: 2014-03-20
Online published: 2014-09-04
Copyright
论文依据DEM地形描述误差(简称Et)的产生机理,在分析现有Et计算模型的基础上,研究建立了顾及DEM格网布设位置的新型Et计算模型,同时以1:5万黄土丘陵地形为例,采用对比分析法揭示了DEM高程插值模型对Et计算结果准确性的影响。实验测试表明:(1)模型能有效地解算出Et的标准差、平均值、最大值、最小值等指标,准确展示出Et的空间分布特征,有助于实现DEM地形描述质量与应用不确定性的分区评价;(2)与双线性、三次卷积、局部二次多项式等常用DEM插值模型相比,以4×4 DEM格网单元为搜索圆的完全规则样条函数插值模型所重构的DEM地表形态,能更为理想地反映Et的量值大小和空间分布。
王春 , 顾留碗 , 陶旸 , 刘玉婵 . DEM地形描述误差(Et)计算模型研究[J]. 地球信息科学学报, 2014 , 16(5) : 699 -706 . DOI: 10.3724/SP.J.1047.2014.00699
The DEM Terrain Description Error (Et) has a basic fact that even if the elevation sampling errors of DEM grid points are equal to zero, a terrain description error will still exist because of DEM simple matrix data format. With the development of earth observation technology, Et has become the main factor affecting the quality of DEM, especially when the DEM grid resolution is low. Traditional calculation models could not calculate the Et value of random points on terrain surface. A new calculation model of DEM terrain description error is presented in this paper for improving the traditional models. The new Et calculation model of any ground point considers the layout position of DEM grid which is established based on the mechanism of Et production. Statistic indexes of Et based on DEM are firstly presented. Then the layout scheme of DEM grid on different datum mark position is designed. At last, the Et value of any ground point considering the layout position of DEM grid is calculated using the new model of Et. 100 DEMs with 5、15、25、50、75、100 and 150 meters resolution for Loess hilly test area in Shannxi Province. The results show that: (1) the standard deviation, average value, extreme value, extreme difference and other metrics indexes of Et for each ground point in any DEM grid resolution of different physiognomy type areas could be calculated scientifically and effectively; (2) reconstruction methods of terrain surface based on DEM strongly influence the accuracy of Et model. Complete regularized spline function interpolation method based on a 4×4 search circle is tested for the optimal interpolation method.
Key words: DEM; terrain description error; morphology precision; cell size; quality analysis
Fig.2 An illustration of different DEM analysis windows图2 栅格分析窗口示意图 |
Fig.3 The location relationship between ground point P and DEM grid图3 地面点P与DEM格网间的位置关系图 |
Tab.1 The statistics indexes of DEM Et表1 DEM Et 数理统计指标 |
序号 | 名称 | 计算公式 | 指标作用 |
---|---|---|---|
1 | 标准差 | 描述Et的聚集离散程度 | |
2 | 平均值 | 描述Et的集中趋势。如果平均值大于0,表明DEM地形曲面存在普遍高于实际地形曲面的系统误差,反之存在低于实际地形曲面的系统误差 | |
3 | 最大值 | 描述DEM地形曲面点高于实际地形曲面的最大偏差 | |
4 | 最小值 | 描述DEM地形曲面点低于实际地形曲面的最大偏差 |
注:式中n为不同格网点布设位置但相同格网分辨率的DEM数量 |
Fig.4 The DEM grid layout scheme of different datum mark positions图4 不同基准点位置的DEM格网布设方案示意图 |
Fig.5 Elevation error distribution of 5m DEM which is reconstructed based on 25m DEM图5 25 m DEM不同插值方法重构的5 m DEM高程数值误差与误差极值分布图 |
Fig.6 The diagram of 4×4 DEM grid analysis window图6 4×4 DEM格网分析窗口示意图 |
Fig.7 The spatial distribution maps of DEM Et in the test area (scale:1:50 000)图7 实验样区DEM Et空间分布特征示意图(比例尺:1:5万) |
Fig.8 The spatial distribution maps of traditional models for calculating Et value (scale:1:50 000)图8 传统模型解算的Et值空间分布特征示意图(比例尺:1:5万) |
Fig.9 Sketch map of Et shell图9 Et壳示意图 |
The authors have declared that no competing interests exist.
[1] |
|
[2] |
|
[3] |
|
[4] |
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
[22] |
|
/
〈 | 〉 |