摘要
准确获取城市不透水面信息对洪涝治理、生态保护以及智慧城市建设等具有重要意义。该文基于Google Earth Engine云平台,以Landsat TM/ETM+/OLI影像为数据源,重构目标年份的最佳影像数据集;在此基础上构建光谱和纹理的分类特征集,采用随机森林分类算法提取不透水面信息,并分析湖南省1987-2017年每3年的不透水面时空变化特征。结果表明:1)该文方法能够有效提取不透水面信息,分类总体精度和Kappa系数分别为97.19%和0.92。2)湖南省不透水面扩张显著,面积从1987年的1107.48 km^2增至2017年的3929.26 km^2,平均扩展速率为94.06 km^2/a,动态度为8.49%。增长较快区域集中在长沙(19.14 km^2/a)、常德(8.46 km^2/a)、衡阳(8.03 km^2/a)、岳阳(7.10 km^2/a)、株洲(6.50 km^2/a)等地;长株潭城市群的不透水面扩张显著,扩展速率和动态度分别为30.28 km^2/a和14.27%。
Accurately monitoring the information of impervious surface in Hunan Province is of great significance for flood control,ecological protection and smart city construction.Based on the Google Earth Engine platform(GEE),using the Landsat TM/ETM+/OLI images as the main data sources to generate the best image dataset with lowest cloud and largest NDVI composite,the impervious surface information of Hunan Province every three years from 1987 to 2017 is extracted and its spatiotemporal variation characteristics are analyzed by classifying feature set from spectral and texture combining with the random forest classifier.The results are as follows.1)Our method can be effectively applied to extract the impervious surface information with an average overall accuracy(OA)of 97.19%and a Kappa coefficient of 0.92.2)During the thirty years,the impervious surface expansion of Hunan Province has increased significantly from 1107.48 km^2 in 1987 to 3929.26 km^2 in 2017 with an average annual expansion rate of 94.06 km^2/a and a dynamic degree of 8.49%.The areas with rapid expansion of impervious surface in Hunan Province are mainly concentrated in Changsha(19.14 km^2/a),Changde(8.46 km^2/a),Hengyang(8.03 km^2/a),Yueyang(7.10 km^2/a)and Zhuzhou(6.50 km^2/a).The impervious surface expansion of the Chang-Zhu-Tan urban agglomeration is significant,with an expansion rate and dynamic degree of 30.28 km^2/a and 14.27%,respectively.
作者
唐志光
邓刚
李少为
李朝奎
陈浩
彭焕华
TANG Zhi-guang;DENG Gang;LI Shao-wei;LI Chao-kui;CHEN Hao;PENG Huan-hua(National-Local Joint Engineering Laboratory of Geo-spatial Information Technology,Hunan University of Science and Technology,Xiangtan 411201;School of Resourse,Environment and Safety Engineering,Hunan University of Science and Technology,Xiangtan 411201;School of Geosciences and Info-physics,Central South University,Changsha 410083,China)
出处
《地理与地理信息科学》
CSCD
北大核心
2020年第2期41-46,共6页
Geography and Geo-Information Science
基金
国家自然科学基金项目(41871058)
湖南省自然科学基金项目(2018JJ3154)
湖南省教育厅科研项目(16C0633)
湖南科技大学校级科研项目(CXTD003)。
