徐晓桃,韩涛,颉耀文.基于单时相MODIS数据的土地覆盖三种分类方法对比研究[J].干旱地区农业研究,2008,(3):253~258 |
基于单时相MODIS数据的土地覆盖三种分类方法对比研究 |
Comparison of the land cover classification methods based on single-temporal MODIS data |
|
DOI:10.7606/j.issn.1000-7601.2008.03.50 |
中文关键词: MODIS 最大似然法 BP神经网络 决策树 See5.0 土地覆盖分类 |
英文关键词:MODIS maximum likelihood classifier BP neural network decision tree classifier See 5.0 land cover classification |
基金项目:甘肃省自然科学基金项目(3ZS051-A25-010); 中国气象局气象新技术推广项目(CMATG2007Z09); 甘肃省气象局科研项目(2008-08); 甘肃省人工增雨防雹效果检验与评估研究(2002人影-2) |
|
摘要点击次数: 180 |
全文下载次数: 144 |
中文摘要: |
以甘肃省为试验区,基于单时相MODIS数据,主要利用其可见光多波段光谱信息,分别使用最大似然法、BP神经网络算法以及基于See 5.0数据挖掘的决策树分类方法对土地覆盖进行了自动分类研究,结果验证表明:决策树分类性能最优,总分类精度达到82.13%,神经网络算法次之,总分类精度为77.60%,最大似然法最差,总分类精度为73.93%;加入boosting技术的See 5.0数据挖掘决策树方法能够快速地进行决策树的建立且能很好地提高较难识别地物类型的分类精度。 |
英文摘要: |
Based on single-temporal MODIS data of Gansu Province mainly using its visible spectra three classifiers-the maximum likelihood BP neural network and decision tree based on data mining software of See 5.0 are used for land cover classification research.The validated result shows that the decision tree algorithm has the best performance of extraction with an overall accuracy of 82.13% followed by the BP network algorithm and the maximum likelihood classifier has the worst performance.Data mining software of See 5.0 with boosting technique can build decision tree quickly and improve the precision of miscible classes. |
查看全文 查看/发表评论 下载PDF阅读器 |