摘要
为了实现章贡区土地信息提取,给城市土地分类提供方法上的参考,并能在此基础上了解到章贡区建成区用地规模变化趋势,以赣州市章贡区为例,将2001年、2004年、2006年、2009年、2012年、2014年这6个年份下同期遥感影像图作为数据源,运用决策树分类法进行提取分类,并运用改进后灰色经济计量模型对未来章贡区建成区用地规模加以预测.研究结果表明,运用决策树分类法提取后的分类精度较高;运用改进后灰色经济计量模型预测得到2015年、2016年、2017年这3年的章贡区建成区用地面积分别为135.59 km2,154.43 km2,173.26 km2.将固有灰色经济计量模型进行改进,得到以修正之后的拟合数据作为实验初始数据建立模型,对比预测表明预测结果得到了改善.
In order to achieve the land information extraction of Zhanggong District in Ganzhou city, make references for method to urban land classification, and further understand the changing trends of Zhanggong built-up areas, the study takes Zhanggong District as the research object, and the remote sensing images in the years of 2001, 2004, 2006, 2009, 2012, 2014 as the data source. Then the decision tree classification method is used in extraction classification and the improved gray econometric model is used in forecasting the future Zhanggong built-up areas. The results show that classification accuracy is higher after extraction by the decision tree classification method; after using the improved gray econometric model, the Zhanggong built-up areas in the year 2015, 2016 and 2017 are predicted to cover 135.59 km^2, 154.43 km^2 and 173.26 km^2. When the gray econometric model is improved, and the corrected fitting data is taken as the experimental model, the predicted results show improvement.
出处
《江西理工大学学报》
CAS
2015年第5期55-60,共6页
Journal of Jiangxi University of Science and Technology
基金
江西省教育厅青年基金项目(GJJ13377)
关键词
决策树分类法
改进后灰色经济计量模型
预测
decision tree classification method
improved gray econometric model
prediction