中文摘要：

以4种MODIS时序数据为基础,以湖北省为研究区,开展区域森林类型快速提取的分类方法研究。将湖北省的森林植被划分为针叶林、阔叶林、混交林、竹林、灌木林和非林地6种地类,通过分析不同森林类型一年内的生长差异,根据各植被指数的均值建立了研究区各森林类型的分类参数,选取2010年MODIS数据产品NDVI第10期、NDVI第12期、NDVI第15期、EVI第10期、LAI第16期、LAI第17期数据,建立了针叶林、阔叶林、混交林、竹林、灌木林、非林地的决策树分类模型,并对研究区进行了分类。结果表明：植被指数均值曲线具有一定的区域地类代表性,分类参数对不同森林类型具有较强的分异性;将分类结果与第八次全国森林资源清查结果相比较,森林类型分类总体精度为85.45%,KAPPA系数为0.770 1,效果较好,说明通过MODIS数据进行区域的森林类型提取是可行的。

英文摘要：

Based on the four MODIS time series data, get around the class when the vegetation index 2010 trends and identify vegetation index greater degree of separation phase, taking Hubei province as the studying area to carry out regional forest type quick extraction,Implementation of the main forest type classification.Divida the forest vegetation of Hubei province for eonifereus forest, broad-leaved forest, mixed forest, bamboo forest, shrub, and non-forest of six types.Through the analysis of different forest types within one year of growth difference. According to the average of each vegetation index established classification parameters of different forest types in the study area, Select the first I0 multi-temporal NDVI, NDVI Article 12, Article 15 NDVI, EVI section 10, LAI article 16, LAI data section 17, established the coniferous forest, broad-leaved forest, mixed forest, bamboo forest, shrub, non-forest＇ s decision tree model, achieved the identification of the forest type information.Research showed that the vegetation index average curve in this paper has certain representative areas, the classification parameters are different for different forest types.The results compared with the result of the eighth national forest resource inventory, the overall classification accuracy of forest types was 85.45%,KAPPA coefficient is 0.770 1, it shows that it is feasible to extract the forest information through MODIS data.