| 基于多方位DEM地形晕渲的黄土地貌正负地形提取 |
| |
| 引用本文： | 陈永刚,汤国安,周毅,李发源,宴实江,张磊.基于多方位DEM地形晕渲的黄土地貌正负地形提取[J].地理科学,2012(1):105-109. |
| |
| 作者姓名： | 陈永刚 汤国安 周毅 李发源 宴实江 张磊 |
| |
| 作者单位： | 南京师范大学虚拟地理环境教育部重点实验室;浙江农林大学浙江省森林生态系统碳循环与固碳减排重点实验室 |
| |
| 基金项目： | 国家自然科学基金项目(40930531、40801148);浙江省教育厅项目(Y201017891)资助 |
| |
| 摘 要： | 以陕北绥德县韭园沟流域5 m分辨率DEM数据为基础,在数字地形分析、多元统计和数据挖掘方法的基础上,提出利用多方位DEM地形晕渲、坡度等多元指标,以主成分分析消除多重共线性和约减维数,并以Logis-tic回归模型提取黄土高原正、负地形的方法。研究结果表明:模型提取精度为82.1%,Kappa统计量为0.629;模型在6个不同流域测试样本上正、负地形的平均精度分别为77.6%,84.9%,加权平均精度为81.3%,模型具有较好的一致性和泛化能力,正、负地形提取效果良好。
|
| 关 键 词： | 正负地形 黄土高原 主成分分析 Logistic回归 |
The Positive and Negative Terrain of Loess Plateau Extraction Based on the Multi-azimuth DEM Shaded Relief |
| |
| CHEN Yong-gang,TANG Guo-an,ZHOU Yi,LI Fa-yuan,YAN Shi-jiang,ZHANG Lei.The Positive and Negative Terrain of Loess Plateau Extraction Based on the Multi-azimuth DEM Shaded Relief[J].Scientia Geographica Sinica,2012(1):105-109. |
| |
| Authors: | CHEN Yong-gang TANG Guo-an ZHOU Yi LI Fa-yuan YAN Shi-jiang ZHANG Lei |
| |
| Institution: | 1(1.Key Laboratory of Virtual Geogrpahic Environment of Ministry of Education,Nanjing Normal University, Nanjing,Jiangsu 210046,China;2.Zhejiang Provincial Key Laboratory of Carbon Cycling in Forest Ecosystems and Carbon Sequestration,Zhejiang Agriculture and Forestry University,Lin’an,Zhejiang 311300,China) |
| |
| Abstract: | Based on the DEM data of Jiuyuangou wateshed in Suide County,Shaanxi of China,with a spatial resolution of 5 m,employing the digital terrain analysis,multivariate statistics and data mining methods,multiple indexes of multi-azimuth DEM shaded relief and slope are established.The solutions of extraction positive and negative terrain of loess plateau by Principle Components Analysis,and Logistic regression model is proposed.The result indicates that: the extracting model has better consistency and accuracy,of which the accuracy is 82.1% and Kappa statistics is 0.6298.Tested on 6 samples of different valley by this model,the mean accuracy of positive and negative terrain are at 77.6% and 84.9%,and weighed mean accuracy is 81.3%.It is suitable to extract positive and negative terrain of loess plateau by PCA and Logistic regression model. |
| |
| Keywords: | positive and negative terrain Loess Plateau Principle Components Analysis Logistic regression |
| 本文献已被 CNKI 等数据库收录！ |
|
