城市土壤钴含量高光谱反演  

Hyperspectral Inversion of Cobalt Content in Urban Soils

作　　者：钟晴 麦麦提吐尔逊·艾则孜[1,2] 米热古力·艾尼瓦尔 侯茂蕤 李浩然 ZHONG Qing;Mamattursun EZIZ;Mireguli AINIWAER;HOU Mao-rui;LI Hao-ran(College of Geographical Science and Tourism,Xinjiang Normal University,U rumqi 830054,China;Laboratory of Arid Zone Lake Environment and Resources,Xinjiang Normal Uni versity,Urumqi 830054,China;College of Computer Science and Technology,Tiangong University,Tianjin 300380,China;School of Geography and Tourism,Shaanxi Normal University,Xi an 710119,China)

机构地区：[1]新疆师范大学地理科学与旅游学院,新疆乌鲁木齐830054 [2]新疆干旱区湖泊环境与资源实验室,新疆乌鲁木齐830054 [3]天津工业大学计算机科学与技术学院,天津300380 [4]陕西师范大学地理科学与旅游学院,陕西西安710119

出　　处：《光谱学与光谱分析》2024年第11期3266-3272,共7页Spectroscopy and Spectral Analysis

基　　金：国家自然科学基金项目(U2003301)资助。

摘　　要：钴元素(Co)被国际癌症研究机构(IARC)列为2B类致癌物,对城市生态系统的安全有潜在危害,如何快速、准确检测土壤中Co元素含量尤为重要。高光谱技术对土壤Co含量反演具有极大潜力。在新疆乌鲁木齐市采集表层(0~20 cm)土壤样品88个,测定Co含量和原始光谱反射率,对原始光谱反射率进行预处理和均方根(RMS)、对数(LT)、对数的倒数(RL)、倒数(RT)、倒数的对数(AT)、一阶微分(FD)、二阶微分(SD)、倒数一阶微分(RTFD)、倒数二阶微分(RTSD)、对数一阶微分(LTFD)、对数二阶微分(LTSD)、均方根一阶微分(RMSFD)、均方根二阶微分(RMSSD)、倒数的对数一阶微分(ATFD)、倒数的对数二阶微分(ATSD)、对数的倒数一阶微分(RLFD)和对数的倒数二阶微分(RLSD)等17种变换。将18种形式的土壤反射率光谱值与Co含量进行Pearson相关性分析(PCC)和CARS优化,筛选出用于建模的特征光谱变量。将筛选出的光谱特征变量分别作为自变量,土壤Co含量作为因变量,基于偏最小二乘回归(PLSR)、随机森林回归(RFR)和支持向量机回归(SVMR)三种算法,构建城市土壤Co含量高光谱反演模型,模型的评价指标采用决定系数(R^(2))、均方根误差(RMSE)和平均绝对误差(MAE)。结果表明:城市土壤Co含量的高光谱模型的稳定性和估算精度由高到低依次为RFR模型、PLSR模型、SVMR模型。Co含量的最佳估算模型是ATFD-RFR模型(R^(2)=0.871,RMSE=0.124,MAE=0.273),相较R-RFR模型R^(2)提高了0.335,RMSE减少了0.32,MAE减少了0.243,RPD为7.90。光谱变换可以有效增强光谱特征,一阶微分变换对光谱特征的增强效果最显著,其中,RTFD不仅可以有效增强Co的光谱特征,还可以很好地提高模型的估算精度。在样点空间异质性不显著、实测值低且均匀时,RFR模型可以在绿洲城市土壤高光谱反演估算中推广。Cobalt(Co)was classified as a group 2B carcinogen by the International Agency for Research on Cancer.It is potentially harmful to the safety of the entire urban ecosystem,and it is particularly important to quickly and accurately detect soil Co content.Hyperspectral techniques have great potential for inversion of soil Co content.88 surface(0~20 cm)soil samples were collected from Urumqi,Xinjiang,to determine the Co content and original spectral reflectance.The original spectral reflectance was preprocessed and applied with 17 types of transformation,which include the root-mean-square(RMS),the logarithm of the logarithm(LT),the inverse of the logarithm(RL),the inverse of the logarithm(RT),the logarithm of the inverse(AT),the first-order differentiation(FD),the second-order differentiation(SD),the inverse first-order differentiation(RTFD)(RTSD),logarithmic first-order differentiation(LTFD),logarithmic second-order differentiation(LTSD),root-mean-square first-order differentiation(RMSFD),root-mean-square second-order differentiation(RMSSD),logarithmic first-order differentiation of the inverse(ATFD),logarithmic second-order differentiation of the inverse(ATSD),logarithmic first-order differentiation of the inverse(RLFD)and logarithmic second-order differentiation(RLSD).Then,the Co content and 18 types of soil spectral data were subjected to Pearson correlation analysis(PCC)and CARS to screen the spectral signature variables for modeling.The soil Co content was taken as the dependent variable,and the screened spectral feature variables were taken as independent variables.Based on three algorithms,namely partial least squares regression(PLSR),random forest regression(RFR),and support vector machine regression(SVMR),the hyperspectral inversion models of urban soil Co content were constructed,and the coefficient of determination(R^(2)),the root-mean-square error(RMSE)and the mean absolute error(MAE)were used as the evaluation indexes.Some conclusions can be drawn:The hyperspectral models'estimation accuracy and stabili

关 键 词：高光谱 钴(Co)元素 反演模型 乌鲁木齐 城市土壤 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]