基于珞珈一号夜光遥感的城市建成区提取方法对比分析研究——以南昌市为例  

Comparative Analysis of Urban Built-up Area Extraction Methods Based on Luojia 1-01 Nighttime Light Remote Sensing: A Case Study of Nanchang

作　　者：王毓乾[1,2,3] 宋雪鹏 邓志杰 李岳 谢相建 程朋根[1,2] WANG Yuqian;SONG Xuepeng;DENG Zhijie;LI Yue;XIE Xiangjian;CHENG Penggen(Key Laboratory of Mine Environmental Monitoring and Improving around Poyang Lake,Ministry of Natural Resources,East China University of Technology,Nanchang 330013,China;School of Geomatics,East China University of Technology,Nanchang 330013,China;Key Laboratory for Digital Land and Resources of Jiangxi Province,East China University of Technology,Nanchang 330013,China;Gongqing Institute of Science and Technology,Jiujiang 332020,China)

机构地区：[1]东华理工大学自然资源部环鄱阳湖区域矿山环境监测与治理重点实验室,江西南昌330013 [2]东华理工大学测绘工程学院,江西南昌330013 [3]东华理工大学江西省数字国土重点实验室,江西南昌330013 [4]共青科技职业学院,江西九江332020

出　　处：《东华理工大学学报（自然科学版）》2023年第1期61-67,共7页Journal of East China University of Technology(Natural Science)

基　　金：国家自然科学基金项目(41861052);江西省自然科学基金项目(20202BABL202045,20202BABL213029);江西省数字国土重点实验室开放研究基金项目(DLLJ201915)。

摘　　要：珞珈一号卫星夜光遥感数据作为我国独立研发的新型夜间灯光数据源,具有较高的空间和辐射分辨率。目前缺少利用珞珈一号卫星夜光遥感数据进行城市建成区提取多种方法的综合性能评价。以南昌市为例,选取突变检测、人居指数和支持向量机分类3种方法提取城市建成区,并进行对比分析。结果表明,运用3种方法从珞珈一号数据提取的建成区都具有丰富的空间结构特征,但3种建成区提取方法各有优劣:人居指数法精度最高,但是建成区非常破碎;突变检测法精度最低,但是能够较好地保留城市的整体形状和边界特征;支持向量机分类法精度次高,但建成区也非常破碎。As a newnight time light data source, Luojia 1-01, independently developed by China, has higher spatial and radiation resolution. At present, there is a lack of comprehensive performance evaluation on Luojia 1-01 data by the existing built-up areas extraction methods in the use of nighttime light remote sensing. Nanchang was chosen as the study area. Three methods including mutation detection, habitat index, and support vector machine classification were selected to extract urban built-up areas by using Luojia 1-01 data. Then their applicability on Luojia 1-01 nighttime-light image were compared and analyzed. The results showed that the built-up area extracted by Luojia 1-01 had rich spatial structure characteristics. Three built-up areas extraction methods had their own advantages and disadvantages. Human settle index was of the highest extraction precision, but its built-up area was very broken. The mutation detection method had the lowest accuracy, but it could better retain the overall shape and boundary characteristics. The SVM classification method had the second-highest accuracy, and its built-up area was also very broken.

关 键 词：夜光遥感 珞珈一号 建成区提取 建成区形态 南昌市 

分 类 号：P237[天文地球—摄影测量与遥感]