Research on the Application of Decision Tree in the Classification of Shandong Peninsula Land Use and Land Cover Change

Land use type is the main content of land use and land cover studies, and the Land Dynamic Monitoring has been focusing on getting information of land cover on the global and regional scale quickly and accurately. This paper takes Shandong Peninsula as an example, MODIS remote sensing as data source, and selects MODIS09Q1 products of the first period in 8-day. Such group index number as NDVI, ENI, NDWI, NDMI will be illustrated to analyze the spectral relation between different land use types and construct new wave bands, finally the application of the decision tree in Shandong Peninsula land use and vegetation cover status will be researched. The results show that the adoption of MODIS data in macroscopical land use classification can achieve high classification accuracy. Such land types as the cultivated land and urban residential sites, which distribute in complete range have a higher accuracy, while such scattered land types as rural residential land and grassland have a lower accuracy.