基于支持向量机的某平原河网水质遥感反演研究  

Research on Remote Sensing Inversion of Water Quality in a Plain River Network Based on Support Vector Machine

作　　者：张昂 张艳军[2] ZHANG Ang;ZHANG Yanjun(Power China Zhongnan Engineering Corporation Limited,Changsha 410014,China;School of Water Resources and Hydropower Engineering,Wuhan University,Wuhan 430072,China)

机构地区：[1]中国电建集团中南勘测设计研究院有限公司,湖南长沙410014 [2]武汉大学水利水电学院,湖北武汉430072

出　　处：《云南水力发电》2025年第3期22-24,共3页Yunnan Water Power

摘　　要：利用WorldView-Ⅱ遥感影像和有限的准同步实地监测数据,结合水体组分的光谱特征,建立了某平原河水质反演的支持向量机模型。通过遥感影像反演了溶解氧(DO)、氨氮(NH3-N)、总磷(TP)、总氮(TN)和总有机碳(TOC)的浓度,反演精度较高,相对误差基本在20%以下,同时分析了支持向量机模型误差的来源及改进措施。将模型用于2008-12-20该平原河网水体,所得结果较客观地反应了水质参数的空间分布规律,表明了利用遥感影像进行城市内河水质监测的可行性和应用前景。A support vector machine model for water quality inversion in a certain plain was established using WorldView-II remote sensing images and limited quasi synchronous field monitoring data,combined with the spectral characteristics of water components.The concentrations of dissolved oxygen(DO),ammonia nitrogen(NH3-N),total phosphorus(TP),total nitrogen(TN),and total organic carbon(TOC)were inverted through remote sensing images,with high inversion accuracy and relative errors below 20%.At the same time,the sources of support vector machine model errors and improvement measures were analyzed.The model was applied to the plain river network water body on December 20,2008,and the results objectively reflected the spatial distribution of water quality parameters,indicating the feasibility and application prospects of using remote sensing images for urban inland water quality monitoring.

关 键 词：WorldView-Ⅱ遥感影像 水质反演 支持向量机模型 

分 类 号：TU461[建筑科学—结构工程] X832[建筑科学—土工工程]