摘要
黑臭水体的遥感识别对水环境监测治理具有重要意义。文章对廊坊市及其周边区域地面采集水体数据的光谱信息进行了分析,在此基础上提出基于决策树的黑臭水体识别算法,并利用“高分二号”(GF-2)卫星影像进行验证,分析黑臭水体空间分布特点及成因。利用决策树模型对GF-2卫星2020年5月影像黑臭水体空间分布进行排查识别,研究结果表明:1)通过卫星同步样点验证,决策树算法识别准确率达到80%,在迁移实验中也有79%准确率,优于传统的阈值法与波段指数法,能有效地识别出黑臭水体;2)黑臭水体分布广泛且不连续,水体受到工业生产、人类生活影响较大,窄小河道、坑塘和断头浜容易发生黑臭现象。利用文中所提方法进行黑臭水体遥感识别,精度高且响应迅速,在水环境管理监测过程中具有较高的应用价值。
The remote sensing recognition of black and odorous water bodies is of great significance to water environment monitoring and management.Through ground data collection in Langfang and its surrounding areas,a black and odorous water body recognition algorithm based on decision trees is proposed and GF-2 satellite images are used.Perform verification and analyze the spatial distribution characteristics and causes of black and odorous water bodies.The decision tree model checked and identified the spatial distribution of black and smelly water bodies on the May 2020 image of the GF-2 satellite,the research results show that:1)Through the verification of satellite synchronization samples,the decision tree algorithm has an accuracy of 80%,and it has an accuracy of 79%in the migration experiment,which is better than the traditional threshold method and band index method,and can effectively identify black Smelly water bodies;2)Black and smelly water bodies are widely distributed and discontinuous.The water bodies are subject to industrial production and have a greater impact on human life.Narrow rivers,pit ponds and broken end creek are prone to black and odorous phenomena.The remote sensing recognition of black and odorous water bodies using this method has high accuracy and rapid response,which has important application value in the process of water environment management and monitoring.
作者
韩文龙
赵起超
金永涛
苑林
罗巍
HAN Wenlong;ZHAO Qichao;JIN Yongtao;YUAN Lin;LUO Wei(North China Institute of Aerospace Engineering,School of Remote Sensing Information Engineering,Langfang 065000,China;Collaborative Innovation Center of Aerospace Remote Sensing Information Processing and Application of Hebei Province,Langfang 065000,China;Ecological Environment Bureau of Langfang City,Hebei Province,Langfang 065600,China)
出处
《航天返回与遥感》
CSCD
北大核心
2022年第1期120-128,共9页
Spacecraft Recovery & Remote Sensing
基金
国防基础科研计划(JCKY201904D004)
河北省自然科学基金(2020409005)
河北省全职引进高端人才科研项目(2020HBQZYC002)。
关键词
黑臭水体
“高分二号”卫星影像
决策树
遥感应用
black and odorous water body
GF-2 image
decision tree
remote sensing applications
作者简介
韩文龙,男,1997年生,北华航天工业学院航空宇航科学与技术专业在读硕士研究生。研究方向为遥感图像解译与地物识别。E-mail:m17803269306@163.com;通讯作者:赵起超,男,1987年生,北华航天工业学院遥感信息工程学院讲师,主要从事水环境遥感监测方向研究。E-mail:theoddone1987@163.com。
