基于高光谱的土壤重金属污染监测方法研究  

Detection of Heavy Metal Pollution in Soil based on Hyperspectral

作　　者：雷晓慧 楚君 Lei Xiaohui;Chu Jun(Henan Shangqiu Ecological Environment Monitoring Center,Shangqiu 476000,China)

机构地区：[1]河南省商丘生态环境监测中心,河南商丘476000

出　　处：《环境科学与管理》2023年第3期108-112,共5页Environmental Science and Management

摘　　要：土壤重金属污染会对农作物和自然植物的生长构成严重威胁。设计高效率的土壤重金属低浓度污染检测方法,对于土壤环境保护至关重要。为此,提出一种基于高光谱的土壤重金属污染监测方法。首先,利用机载光谱扫描仪远程采集土壤的高光谱样本,并建立土壤样本的高光谱数据集。然后,以数据集为基础,分析土壤中重金属元素的生物特征,提取高光谱数据特征。最后,利用K-means聚类算法对特征数据实施处理,判断土壤重金属污染的类型,实现高精度监测。实验结果表明,该方法的监测准确度较高,且监测性能不受重金属污染浓度的限制。Heavy metal pollution in soil will pose a serious threat to the growth of crops and natural plants.It is very important for soil environmental protection to design an efficient detection method for low concentration pollution of heavy metals in soil.Therefore,a hyperspectral method for monitoring heavy metal pollution in soil was proposed.Firstly,the hyperspectral samples of soil are collected remotely by airborne spectral scanner,and the hyperspectral data set of soil samples is established.Then,based on the data set,the biological characteristics of heavy metal elements in soil are analyzed and the hyperspectral data characteristics are extracted.Finally,K-means clustering algorithm is used to process the characteristic data,judge the type of heavy metal pollution in soil,and realize high-precision monitoring.The experimental results show that the monitoring accuracy of this method is high,and the monitoring performance is not limited by the concentration of heavy metal pollution.

关 键 词：机载光谱扫描仪 高光谱数据 自然环境污染源 K-MEANS聚类算法 

分 类 号：X830.2[环境科学与工程—环境工程]