基于单时相MODIS数据的决策树自动构建及分类研究

¹中国气象局兰州干旱气象研究所,兰州市东岗东路207号730020;²甘肃省干旱气候变化与减灾重点实验室,兰州市东岗东路207号,730020;³中国气象局干旱气候变化与减灾重点开放实验室,兰州市东岗东路207号,730020;⁴兰州大学西部环境教育部重点实验室,兰州市天水南路222号730000

基金项目: 甘肃省自然科学基金资助项目(3ZS051-A25-010)；甘肃省气象局科研基金资助项目(2008-08,2002人影-2)；中国气象局气象新技术推广计划资助项目(CMATG2007Z09)；中国气象局风云三号科研基金资助项目

详细信息

Automated Construction and Classification of Decision Tree Classifier Based on Single-Temporal MODIS Data

¹Institute of Arid Meteorology,CMA,2070 East Donggang Road,Lanzhou 730020,China;²Key Laboratory of Arid Climatic Change and Reducing Disaster of Gansu Province,2070 East Donggang Road,Lanzhou 730020,China;³Key Open Laboratory of Arid Climatic Change and Reducing Disaster,CMA,2070 East Donggang Road,Lanzhou 730020,China;⁴Key Laboratory of Western China's Environment Systems,Ministry of Education,Lanzhou University,222 South Tianshui Road,Lanzhou 730000,China

Funds: 甘肃省自然科学基金资助项目(3ZS051-A25-010)；甘肃省气象局科研基金资助项目(2008-08,2002人影-2)；中国气象局气象新技术推广计划资助项目(CMATG2007Z09)；中国气象局风云三号科研基金资助项目

摘要: 以甘肃省为试验区,利用单时相MODIS数据的光谱信息,使用最大似然法和基于See 5.0数据挖掘的决策树分类方法,进行了分类对比研究。分类结果表明,加入温度-植被角度TVA和温度-植被距离TVD两个指数后,低植被覆盖区的分类效果得到了改善;基于See 5.0数据挖掘的决策树方法能够快速地建立决策树,且能提高较难识别地物类型的分类精度。
- MODIS /
- 最大似然法 /
- 决策树 /
- See5.0 /
- 土地覆盖分类
Abstract: With single-temporal MODIS data of Gansu Province,we used a maximum likelihood method and decision tree methed based on data mining software See 5.0 to stuly the land cover classification.The experimental result shows that the accuracy of low vegetation area is improved with the indexes of TVA and TVD,compared with Maximum likelihood classfier,and data mining software See 5.0 is able to build decision tree quickly and improve the precision of miscible classes.
- MODIS /
- maximum likelihood classfier /
- decision tree classifier /
- See 5.0 /
- land cover classification