Study on Rapid Investigation of Heavy Metal Pollution in Site Soil Based on Hyperspectral Technology
In order to accurately predict the distribution of heavy metals in the soil of the site and achieve rapid investigation of soil heavy metal pollution,the soil in the landfill area of a certain waste additive factory is taken as the research object,based on hyperspectral data,univariate regression model,partial least squares regression model,and support vector machine model are used to estimate the content of heavy metals such as Cr,Ni,Cu,Zn,Cd,Pb,As,and Hg in the soil.The results show that there is a negative correlation between soil spectral reflectance and the content of various heavy metals;the partial least squares regression model and support vector machine model have better prediction accuracy for 8 heavy metals than univariate regression model,and the partial least squares regression model is the best estimation model for Cd,Pb,Cr,and Ni,while the support vector machine model is the best prediction model for Cu,As,Zn,and Hg;the trend of soil heavy metal inversion results in the research area is basically consistent with the laboratory analysis results,and the distribution of high value areas and extreme points is also relatively consistent,which can delineate areas with heavy metal pollution risks and provide technical support to achieve rapid investigation of soil heavy metal pollution on the site.
soilheavy metalshyperspectral reflectancepartial least squares regression modelsupport vector machine model
国家科技期刊平台
NSTL
万方数据
