基于Sentinel-1/2的水体提取方法对比研究——以斯里兰卡小型水体为例  

Comparative study on water extraction methods based on Sentinel-1/2:a case study of small water body in Sri Lanka

作　　者：周晗 叶虎平[1,3,4] 魏显虎 张宗科[1,3] 王法溧 齐永菊[1,3] 马建如 ZHOU Han;YE Huping;WEI Xianhu;ZHANG Zongke;WANG Fali;QI Yongju;MA Jianru(Institute of Remote Sensing and Digital Earth,Chinese Academy of Sciences,Beijing 100101,China;University of Chinese Academy of Sciences,Beijing 100049,China;China-Sri Lanka Joint Research and Demonstration Center for Water Technology,Chinese Academy of Sciences,Beijing 100085,China;State Key Laboratory of Resources and Enviormental Information System,Institute of Geographic Sciences and Natural Resources Research,Chinese Academy of Sciences,Beijing 100101,China)

机构地区：[1]中国科学院遥感与数字地球研究所,北京100101 [2]中国科学院大学,北京100049 [3]中国科学院中-斯水技术联合研究中心,北京100085 [4]中国科学院地理科学与资源研究所资源与环境信息系统国家重点实验室,北京100101

出　　处：《中国科学院大学学报（中英文）》2019年第6期794-802,共9页Journal of University of Chinese Academy of Sciences

基　　金：中国科学院战略性先导科技专项(XDA2003030201),中国科学院中-斯水技术研究与示范联合中心项目,中国科学院对外合作重点项目(Y625151401),中国科学院境外机构建设项目(SAJC201608)资助

摘　　要：地表水面精确提取是研究地表水质和水量变化的重要基础。斯里兰卡是"21世纪海上丝绸之路"的重要参与国,年降雨量丰富,但时空分布不均,斯里兰卡人民长期用水困难,研究斯里兰卡地表水体有助于斯里兰卡民生问题的解决。斯里兰卡国内散布着大量的小型水库和坑塘,这些小面积水体易受周边环境因素影响而提取困难。基于2017年7月斯里兰卡中东部地区的哨兵(Sentinel)1/2号卫星影像,对比分析单波段法、水体指数法和监督分类等水体提取方法的精度和存在的问题。结果表明,归一化水体指数法NDWI的准确率最高,分类精度达94%。The accurate extraction of surface water is an important basis for studying the changes of quality and quantity for surface water. Sri Lanka is one of the most important participating countries along "21 st-Century Maritime Silk Road".Rainfall is abundant in Sri Lanka. However, due to uneven spatial and temporal distributions of the rainfall, people in Sri Lanka have a long-term difficulty in using water. Study on surface water is helpful for solving the problem of people’s livelihood in Sri Lanka. There are a large number of small reservoirs and ponds scattered throughout the country,but it is difficult to extract small area water which is affected by surrounding environment. In this paper, the accuracies and existing problems of water extraction methods, such as single-band method, water body index method, and supervised classification method, were compared and analyzed based on Sentinel 1/2 satellite image in July 2017 in eastern Sri Lanka. The analysis results show that the normalized difference water index(NDWI) method has the highest accuracy with a classification accuracy of 94%.

关 键 词：Sentinel-1/2 水体提取方法 NDWI 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]