SRTM数据精度检测
摘要
航天飞机雷达地形测绘任务数字高程模型是目前能够免费获取的全球范围的DEM数据,已经得到了广泛的应用。但是,对SRTM DEM的数据质量特征等方面的研究还存在一定的盲区。本文首先对我国标准化生产的1：250000比例尺DEM为参考对比数据,选取福建中部地区为实验样区,根据残差从总体精度和随机选点两个方面对SRTM DEM进行分析。由于实验样区位于福建,山地面积较大,地形结构复杂,地形起伏度较大,SRTM DEM的分辨率无法细致地刻画地形起伏变化。另外,由于雷达干涉测量本身的缺陷,在一些山体陡峭的地区,山体背坡坡度较大时可能会产生雷达阴影从而造成数据缺失,仅通过内插等方法填补这些空洞点,因此这些数据缺失的地方就可能会出现更大误差。但是SRTM数据对于精度要求并不高的山地生态与水文分析是满足的。论文选取坡度、坡向、平面曲率、剖面曲率,对SRTM DEM在数字地形方面进行定量分析。结果显示,SRTM DEM在坡度、坡向两个参数的总体精度上高于1：250000 DEM,其中坡度损失量在山地区和丘陵区均较大,在平地区较小；坡向损失量在丘陵区和平地地区较大,而在山地地区较小。在剖面曲率和平面曲率两个参数上,SRTM DEM由于分辨率低于1：250000 DEM,其剖面曲率的面积分布更集中在低曲率值上,平面曲率在较低级别处的面积也随之增大。
Shuttle radar topography mission digital elevation model is able to free access to the global DEM data, has been widely used. However, the SRTM DEM data quality characteristics of the research there is still a blind spot. In this article, standardization of production to China 1:250000 scale DEM data for the reference contrast, the central region of Fujian selected as experimental plots, using the error model, the overall accuracy and relative accuracy from both the analysis of the SRTM DEM. As the experimental plot is located in Fujian, mountain larger, complex terrain, terrain with larger, SRTM DEM resolution can not be accurate to describe the topographical features. In addition, since radar interferometry own shortcomings, and in some mountain areas with a steep, mountain slope and may produce radar shadows resulting in missing data, only by interpolation and other methods to fill the empty spots, so these data missing where more errors may occur. But accuracy is not high mountain ecology and hydrological analysis is sufficient. Papers selected slope, aspect, surface curvature, profile curvature, on the SRTM DEM in digital terrain aspects of quantitative analysis. The results showed that, SRTM DEM in the slope and aspect on the overall accuracy of the two parameters was higher than 1:250000 DEM, in which loss slope in the mountain areas and hilly areas are large, flat area in the small; loss in the hilly slope District peaceful area larger, but smaller in mountain areas. Profile curvature and surface curvature in the two parameters, SRTM DEM as resolution of less than 1:250000 DEM, the profile curvature of surface area distribution is more concentrated in the low curvature values, the surface curvature at the lower level office area also increased.
Shuttle radar topography mission digital elevation model is able to free access to the global DEM data, has been widely used. However, the SRTM DEM data quality characteristics of the research there is still a blind spot. In this article, standardization of production to China 1:250000 scale DEM data for the reference contrast, the central region of Fujian selected as experimental plots, using the error model, the overall accuracy and relative accuracy from both the analysis of the SRTM DEM. As the experimental plot is located in Fujian, mountain larger, complex terrain, terrain with larger, SRTM DEM resolution can not be accurate to describe the topographical features. In addition, since radar interferometry own shortcomings, and in some mountain areas with a steep, mountain slope and may produce radar shadows resulting in missing data, only by interpolation and other methods to fill the empty spots, so these data missing where more errors may occur. But accuracy is not high mountain ecology and hydrological analysis is sufficient. Papers selected slope, aspect, surface curvature, profile curvature, on the SRTM DEM in digital terrain aspects of quantitative analysis. The results showed that, SRTM DEM in the slope and aspect on the overall accuracy of the two parameters was higher than 1:250000 DEM, in which loss slope in the mountain areas and hilly areas are large, flat area in the small; loss in the hilly slope District peaceful area larger, but smaller in mountain areas. Profile curvature and surface curvature in the two parameters, SRTM DEM as resolution of less than 1:250000 DEM, the profile curvature of surface area distribution is more concentrated in the low curvature values, the surface curvature at the lower level office area also increased.
引文
[1]陈俊勇.对SRTM3和GTOPO30地形数据质量的评估[J].武汉大学学报信息科学版,2005,30(11)：941-943
[2]曹玲玲,张秋文.基于SRTM的数字河网提取及其应用[J].人民长江,2007,38(8)：150-152
[3]陈楠,汤国安等.基于不同比例尺的DEM地形信息比较[J].西北大学学报·自然科学版,2003,33(2)：237-240
[4]程三友,刘少峰,张会平等.大别山构造地貌的DEM初步分析[J].地质力学学报,2005,11(4)：333-340
[5]龚健雅.整体SIS的数据组织与处理[M].武汉：武汉测绘科技大学出版社,1993
[6]胡鹏,吴艳兰,胡海.数字高程模型精度评定的基本理论[J].地球信息科学,2003(3)：64-70
[7]洪顺英,申旭辉,荆凤等.基于SRTM DEM的阿尔泰山构造地貌特征分析[J].国土资源遥感,2007(73)：62-68
[8]荆创利.使用SAR影像生成DEM的方法研究与精度分析[D].西南交通大学,2008
[9]李志林,朱庆.数字高程模型[M].武汉：武汉大学出版社,2000
[10]刘勇.中国西北部高山高原地区SRTM3数据质量评价[J].兰州大学学报·自然科学版,2008,44(6)：1-6
[11]马龙,杜道生.GIS中DEM产品精度的分析和评价[J].测绘信息与工程,2003：4-6
[12]马龙,李颖.从GTOPO30到SRTM DEM精度研究——以西藏为例[J].水土保持通报,2006,26(5)：71-74
[13]舒宁.雷达影像干涉测量原理理[M].武汉：武汉大学出版社,2003(13-19)
[14]汤国安,龚健雅,陈正江等.数字高程模型地形描述精度量化模拟研究[J].测绘学报,2001,30(4)：361-365
[15]汤国安,刘学军,闾国年.数字高程模型及地学分析的原理与方法[M].北京：科学出版社,2005
[16]汤国安,杨昕.ArcGIS地理信息系统空间分析实验教程[M].北京：科学出版社,2006
[17]王光霞,朱长青,史文中等.数字高程模型地形描述精度的研究[J]测绘学报,2004,33(2)：168-173
[18]王培法SRTM DEM在TOPMODEL模型中的可用性分析[J].水土保持研究,2005,12(6)：194-205
[19]邬伦,刘瑜等.地理信息系统——原理、方法和应用[M].北京：科学出版社,2001
[20]闫业超,张树文,岳书平.东北川岗地区SRTM数据质量评价[J].中国科学院研究生院学报,2008(25)：41-45
[21]周启鸣,刘学军.数字地形分析[M].北京：科学出版社,2006
[221张卫柱,刘平芝,熊顺.数字高程模型(DEM)精度评估[J].解放军测绘研究所学报,2002(3)：27-30
[23]张会平,刘少锋,孙亚平等.基于区域地形起伏的获取及应用[J].国土资源遥感,2006(67)：31-35
[24]张艳梅,程晓,李斐等.基于SRTM DDEM的二路重轨DINSAR地震形迹探测—以2001年昆轮山口西M8.1大地震城中为例[J].地震,2005,25(4)：105-111
[25]詹蕾.SRTM DEM的精度评价及其适应性研究[D].南京师范大学,2008
[26]左美蓉.SRTM高程数据及其应用研究[D].中南大学,2009
[27]Ackerman.F.The Accuracy of Digital Terrain Model.Proceeding of 37th Photogrammertric Week[J].University of Stuttgart,1979:113-143
[28]Charles G.brown,Jr., Kamal Sarabandi, Leland E.Pierce. Validation of the Shuttle Radar Topography Mission Height Data.[J] IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,2005, 43(8):1707-1715
[29]Denker H. Evaluation of SRTM3 and GOTOP30 Terrain Data in Germany[J]. Proceeding of GGSM 2004, IAG Porto, Portugal,2004
[30]Dan G. Blumberg. Analysis of large Aeolian(wind-blown)bedforms using the Shuttle Radar Topography Mission(SRTM)digital elevation data[J].Remote Sensing of Environment 2006(100):179-189
[31]Li Zhilin. Variation of the Accrracy of Digitla Terrain Models With Sampling Interval[J].Photogrammetric Record,1992,14(79):113-128
[32]Ludwig,R., Hellwich,O., Strunz,G等Applications of digital elevation models from SAR interferometry for hydrologic modeling[J].Photogrammetrie, Fernerdundung, Geoinformation2000(2): 6-18
[2]曹玲玲,张秋文.基于SRTM的数字河网提取及其应用[J].人民长江,2007,38(8)：150-152
[3]陈楠,汤国安等.基于不同比例尺的DEM地形信息比较[J].西北大学学报·自然科学版,2003,33(2)：237-240
[4]程三友,刘少峰,张会平等.大别山构造地貌的DEM初步分析[J].地质力学学报,2005,11(4)：333-340
[5]龚健雅.整体SIS的数据组织与处理[M].武汉：武汉测绘科技大学出版社,1993
[6]胡鹏,吴艳兰,胡海.数字高程模型精度评定的基本理论[J].地球信息科学,2003(3)：64-70
[7]洪顺英,申旭辉,荆凤等.基于SRTM DEM的阿尔泰山构造地貌特征分析[J].国土资源遥感,2007(73)：62-68
[8]荆创利.使用SAR影像生成DEM的方法研究与精度分析[D].西南交通大学,2008
[9]李志林,朱庆.数字高程模型[M].武汉：武汉大学出版社,2000
[10]刘勇.中国西北部高山高原地区SRTM3数据质量评价[J].兰州大学学报·自然科学版,2008,44(6)：1-6
[11]马龙,杜道生.GIS中DEM产品精度的分析和评价[J].测绘信息与工程,2003：4-6
[12]马龙,李颖.从GTOPO30到SRTM DEM精度研究——以西藏为例[J].水土保持通报,2006,26(5)：71-74
[13]舒宁.雷达影像干涉测量原理理[M].武汉：武汉大学出版社,2003(13-19)
[14]汤国安,龚健雅,陈正江等.数字高程模型地形描述精度量化模拟研究[J].测绘学报,2001,30(4)：361-365
[15]汤国安,刘学军,闾国年.数字高程模型及地学分析的原理与方法[M].北京：科学出版社,2005
[16]汤国安,杨昕.ArcGIS地理信息系统空间分析实验教程[M].北京：科学出版社,2006
[17]王光霞,朱长青,史文中等.数字高程模型地形描述精度的研究[J]测绘学报,2004,33(2)：168-173
[18]王培法SRTM DEM在TOPMODEL模型中的可用性分析[J].水土保持研究,2005,12(6)：194-205
[19]邬伦,刘瑜等.地理信息系统——原理、方法和应用[M].北京：科学出版社,2001
[20]闫业超,张树文,岳书平.东北川岗地区SRTM数据质量评价[J].中国科学院研究生院学报,2008(25)：41-45
[21]周启鸣,刘学军.数字地形分析[M].北京：科学出版社,2006
[221张卫柱,刘平芝,熊顺.数字高程模型(DEM)精度评估[J].解放军测绘研究所学报,2002(3)：27-30
[23]张会平,刘少锋,孙亚平等.基于区域地形起伏的获取及应用[J].国土资源遥感,2006(67)：31-35
[24]张艳梅,程晓,李斐等.基于SRTM DDEM的二路重轨DINSAR地震形迹探测—以2001年昆轮山口西M8.1大地震城中为例[J].地震,2005,25(4)：105-111
[25]詹蕾.SRTM DEM的精度评价及其适应性研究[D].南京师范大学,2008
[26]左美蓉.SRTM高程数据及其应用研究[D].中南大学,2009
[27]Ackerman.F.The Accuracy of Digital Terrain Model.Proceeding of 37th Photogrammertric Week[J].University of Stuttgart,1979:113-143
[28]Charles G.brown,Jr., Kamal Sarabandi, Leland E.Pierce. Validation of the Shuttle Radar Topography Mission Height Data.[J] IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,2005, 43(8):1707-1715
[29]Denker H. Evaluation of SRTM3 and GOTOP30 Terrain Data in Germany[J]. Proceeding of GGSM 2004, IAG Porto, Portugal,2004
[30]Dan G. Blumberg. Analysis of large Aeolian(wind-blown)bedforms using the Shuttle Radar Topography Mission(SRTM)digital elevation data[J].Remote Sensing of Environment 2006(100):179-189
[31]Li Zhilin. Variation of the Accrracy of Digitla Terrain Models With Sampling Interval[J].Photogrammetric Record,1992,14(79):113-128
[32]Ludwig,R., Hellwich,O., Strunz,G等Applications of digital elevation models from SAR interferometry for hydrologic modeling[J].Photogrammetrie, Fernerdundung, Geoinformation2000(2): 6-18