MLR和PLSR的沙壤土盐分含量光谱检测对比研究
|
摘要
为了快速有效检测南疆地区典型土壤(沙壤土)的盐分含量变化,利用光谱仪和电导仪测得南疆阿拉尔市红枣种植区盐渍土近红外高光谱和电导率数据,基于7种不同光谱预处理方法和2种特征波长选择算法,分别建立多元线性回归(MLR)和偏最小二乘回归(PLSR)的土壤盐分监测模型。结果表明:7种预处理方法中,归一化,多元散射,变量标准化和一阶导数能够有效提高土壤盐分的预测模型精度。基于多元逐步回归(SMR)波长选择方法的多元线性回归(SMLR)模型的R_(val)~2> 0. 948 9,RPD> 6. 294 9,RMSEP <0. 435 6;基于连续投影算法(SPA)的多元线性回归(SPAMLR)模型的R_(val)~2> 0. 956 8,RPD> 6. 922 1,RMSEP <0. 361 6,预测结果要优于偏最小二乘回归(PLSR)模型,其中基于归一化处理后的SMLR和SPA-MLR的预测精度最为理想,分别为R_(val)~2=0. 979 2,RPD=9. 907 8,RMSEP=0. 287 6和R_(val)~2=0. 980 5,RPD=10. 50,RMSEP=0. 278 3,而且筛选的特征波长较少。说明归一化是更有效的光谱预处理方法,多元线性回归(MLR)更适合建立南疆典型沙壤土盐分含量的预测模型。
In order to monitor typical soil salt content( sandy loam soil) in South Xinjiang,China quickly and effectively,and to improve the precision of the soil salt content estimation model by removing the noise of soil hyperspectral absorbance,this paper focused on the inversion relationship between soil spectrum and electrical conductivity( EC) by using multiple spectral pretreatment methods,and then the multiple linear regression( MLR) and the partial least squares regression( PLSR) modelling were applied to establish the salt content model based on the hyperspectral analysis technique. The effective and predictive capacities of different models were validated. This study took the typical arid area in South Xinjiang as the research object,obtained the hyperspectral data and EC by using Near-infrared spectrometer( Zolix Gaia Sorter) and conductivity meter( DDS-307),142 soil samples at 0 ～ 20 cm depth were collected and these samples were highly representative for the EC. Seven pretreatment methods were used to pretreat the original spectral data,such as vector normalization( VN),multiplicative scatter correction( MSC),standard normal variate( SNV),moving-average( MA) smoothing,Savitzky-Golay( SG) smoothing,first derivative( 1-Der) and second derivative( 2-Der),then the characteristic wavelengths were extracted with stepwise multiple regression( SMR) and successive projection algorithm( SPA),which were used as input variables of MLR and PLSR modeling. The results showed that the optimum pretreatment methods were VN,MSC,SNV and 1-Der. According to different pretreatments,in the stepwise multiple linear regression( SMLR) prediction model R_(val)~2 was greater than 0. 95,RPD was greater than 6. 2,and RMSEP was less than 0. 44. In the multiple linear regression model based on the successive projections algorithm( SPA-MLR) R_(val)~2 was greater than 0. 96,RPD was greater than6. 9,and RMSEP was less than 0. 36,which were better than those in SMLR. In the partial least squares regression prediction model R_(val)~2 was greater than 0. 88,RPD was greater than 4. 4,RMSEP was less than 0. 62 and in the partial least squares regression model based on the successive projections algorithm( SPA-PLSR) R_(val)~2 was greater than0. 59,RPD was greater than 2. 4,and RMSEP was less than 1. 1 which were less than those in SMLR and SPAMLR. The best prediction of SMLR and SPA-MLR after vector normalization( VN) was as the follows: RMSEP =0. 287 6,R_(val)~2= 0. 979 2,RPD = 9. 907 8 and RMSEP = 0. 278 3,R_(val)~2= 0. 980 5,RPD = 11. 50 which had fewer characteristic wavelengths. So the VN is a best effective pretreatment method. It is more suitable to establish the prediction model of soil conductivity in jujube tree by MLR with PLSR. It could be an important part in future researches how to choose the right algorithm to analyze the soil salt content with the spectral data.
In order to monitor typical soil salt content( sandy loam soil) in South Xinjiang,China quickly and effectively,and to improve the precision of the soil salt content estimation model by removing the noise of soil hyperspectral absorbance,this paper focused on the inversion relationship between soil spectrum and electrical conductivity( EC) by using multiple spectral pretreatment methods,and then the multiple linear regression( MLR) and the partial least squares regression( PLSR) modelling were applied to establish the salt content model based on the hyperspectral analysis technique. The effective and predictive capacities of different models were validated. This study took the typical arid area in South Xinjiang as the research object,obtained the hyperspectral data and EC by using Near-infrared spectrometer( Zolix Gaia Sorter) and conductivity meter( DDS-307),142 soil samples at 0 ～ 20 cm depth were collected and these samples were highly representative for the EC. Seven pretreatment methods were used to pretreat the original spectral data,such as vector normalization( VN),multiplicative scatter correction( MSC),standard normal variate( SNV),moving-average( MA) smoothing,Savitzky-Golay( SG) smoothing,first derivative( 1-Der) and second derivative( 2-Der),then the characteristic wavelengths were extracted with stepwise multiple regression( SMR) and successive projection algorithm( SPA),which were used as input variables of MLR and PLSR modeling. The results showed that the optimum pretreatment methods were VN,MSC,SNV and 1-Der. According to different pretreatments,in the stepwise multiple linear regression( SMLR) prediction model R_(val)~2 was greater than 0. 95,RPD was greater than 6. 2,and RMSEP was less than 0. 44. In the multiple linear regression model based on the successive projections algorithm( SPA-MLR) R_(val)~2 was greater than 0. 96,RPD was greater than6. 9,and RMSEP was less than 0. 36,which were better than those in SMLR. In the partial least squares regression prediction model R_(val)~2 was greater than 0. 88,RPD was greater than 4. 4,RMSEP was less than 0. 62 and in the partial least squares regression model based on the successive projections algorithm( SPA-PLSR) R_(val)~2 was greater than0. 59,RPD was greater than 2. 4,and RMSEP was less than 1. 1 which were less than those in SMLR and SPAMLR. The best prediction of SMLR and SPA-MLR after vector normalization( VN) was as the follows: RMSEP =0. 287 6,R_(val)~2= 0. 979 2,RPD = 9. 907 8 and RMSEP = 0. 278 3,R_(val)~2= 0. 980 5,RPD = 11. 50 which had fewer characteristic wavelengths. So the VN is a best effective pretreatment method. It is more suitable to establish the prediction model of soil conductivity in jujube tree by MLR with PLSR. It could be an important part in future researches how to choose the right algorithm to analyze the soil salt content with the spectral data.
引文
[1] PANG G,TAO W,JIE L,et al. Quantitative model based on fieldderived spectral characteristics to estimate soil salinity in Minqin County,China[J]. Soil Science Society of America Journal,2014,78(2):546-555.
[2] ZHAO Z,TASHPOLAT T,ZHANG F,et al. Spectral characteristics of soil salt content in typical oasis of Tarim River’s middle reaches[J]. Journal of Natural Disasters,2012,21(5):72-78.
[3]吴明珠,李小梅,沙晋明.亚热带红壤全氮的高光谱响应和反演特征研究[J].光谱学与光谱分析,2013,33(11):3111-3115.[WU Mingzhu,LI Xiaomei,SHA Jinming. Spectral inversion models for prediction of red soil total nitrogen content in subtropical region(Fuzhou)[J]. Spectroscopy&Spectral Analysis,2013,33(11):3111-3115.]
[4] NAGY A,PETERRICZU,GALYA B,et al. Spectral estimation of soil water content in visible and near infra-red range[J]. 2014,3(3):163-171.
[5] ZHANG P,LI Y. Study on the comparisons of the establishment of two mathematical modeling methods for soil organic matter content based on spectral reflectance[J]. Spectroscopy&Spectral Analysis,2016,36(3):903-910.
[6]张晓光,黄标,季峻峰,等.基于可见近红外高光谱的东北盐渍土盐分定量模型研究[J].光谱学与光谱分析,2012,32(8):2075-2079.[ZHANG Xiaoguang,HUANG Biao,JI Junfeng,et al. Quantitative prediction of soil salinity content with visible-near infrared hyper-spectra in northeast China[J]. Spectroscopy&Spectral Analysis,2012,32(8):2075-2079.]
[7]杨爱霞,丁建丽,李艳红,等.基于表观电导率与实测光谱的干旱区湿地土壤盐分监测[J].中国沙漠,2016,36(5):1365-1373.[YANG Aixia,DING Jianli,LI Yanhong,et al. Apparent electronic conductivity and measured spectral for monitoring soil salt content in arid lakeside wetland[J]. Journal of Desert Research,2016,36(5):1365-1373.]
[8] GOLDSHLEGER N,CHUDNOVSKY A,BEN-BINYAMIN R. Predicting salinity in tomato using soil reflectance spectra[J]. International Journal of Remote Sensing,2013,34(17):6079-6093.
[9] NETO O R,TEIXEIRA A,LEAO R,et al. Hyperspectral remote sensing for detecting soil salinization using Pro SpecTIR-VS aerial imagery and sensor simulation[J]. Remote Sensing,2017,9(1):1-16.
[10]彭杰,王家强,向红英,等.土壤含盐量与电导率的高光谱反演精度对比研究[J].光谱学与光谱分析,2014,34(2):510-514.[PENG Jie,WANG Jiaqiang,XIANG Hongying,et al. Comparative study on hyperspectral inversion accuracy of soil salt content and electrical conductivity[J]. Spectroscopy and Spectral Analysis,2014,34(2):510-514.]
[11]屈永华,段小亮,高鸿永,等.内蒙古河套灌区土壤盐分光谱定量分析研究[J].光谱学与光谱分析,2009,29(5):1362-1366.[QU Yonghua,DUAN Xiaoliang,GAO Hongyong,et al.Quantitative retrieval of soil salinity using hyperspectral data in the region of Inner Mongolia Hetao Irrigation District[J]. Spectroscopy&Spectral Analysis,2009,29(5):1362-1366.]
[12]丁建丽,伍漫春,刘海霞,等.基于综合高光谱指数的区域土壤盐渍化监测研究[J].光谱学与光谱分析,2012,32(7):1918-1922.[DING Jianli,WU Manchun,LIU Haixia,et al. Study on the soil salinization monitoring based on synthetical hyperspectral index[J]. Spectroscopy&Spectral Analysis,2012,32(7):1918-1922.]
[13]段鹏程,熊黑钢,李荣荣,等.不同干扰程度的盐渍土与其光谱反射特征定量分析[J].光谱学与光谱分析,2017,37(2):571-576.[DUAN Pengcheng,XIONG Heigang,LI Rongrong,et al. A quantitative analysis of the reflectance of the saline soil under different disturbance extent[J]. Spectroscopy&Spectral Analysis,2017,37(2):571-576.]
[14]李亚莉,乔江飞,董天宇,等.不同质地盐渍化土壤水盐含量的高光谱反演[J].应用生态学报,2016,27(12):3807-3815.[LI Yali,QIAO Jiangfei,DONG Tianyu,et al. Hyperspectral inversion of soil water and salt content in soils with different textures[J]. Chinese Journal of Applied Ecology,2016,27(12):3807-3815.]
[15]刘娅,潘贤章,王昌昆,等.基于可见-近红外光谱的滨海盐土土壤盐分预测方法[J].土壤学报,2012,49(4):824-829.[LIU Ya,PAN Xianzhang,WANG Changkun,et al. Prediction of coastal saline soil salinity based on vis-nir reflectance spectroscopy[J]. Acta Pedologica Sinica,2012,49(4):824-829.]
[16] ZHAN X Y,ZHAO N,LIN Z Z,et al. Effect of algorithms for calibration set selection on quantitatively determining asiaticoside content in centella total glucosides by near infrared spectroscopy[J].Spectroscopy&Spectral Analysis,2014,34(12):3267-3272.
[17]尼珍,胡昌勤,冯芳.近红外光谱分析中光谱预处理方法的作用及其发展[J].药物分析杂志,2008,28(5):824-829.[NI Zhen,HU Changqin,FENG Fang. Progress and effect of spectral data pretreatment in NIR analytical technique[J]. Chinese Journal of Pharmaceutical Analysis,2008,28(5):824-829.]
[18]夏俊芳,李培武,李小昱,等.不同预处理对近红外光谱检测脐橙VC含量的影响[J].农业机械学报,2007,38(6):107-111.[XIA Junfang,LI Peiwu,LI Xiaoyu,et al. Effect of different pretreatment method of nondestructive measure vitamin C content of umbilical orange with near-infrared spectroscopy[J]. Transactions of the Chinese Society for Agricultural Machinery,2007,38(6):107-111.]
[19]刘娅,潘贤章,王昌昆,等.土壤湿润条件下基于光谱对称度的盐渍土盐分含量预测[J].光谱学与光谱分析,2013,33(10):2771-2776.[LIU Ya,PAN Xianzhang,WANG Changkun,et al.Predicting soil salinity based on spectral symmetry under wet soil condition[J]. Spectroscopy&Spectral Analysis,2013,33(10):2771-2776.]
[20]陈文倩,丁建丽,谭娇,等.干旱区绿洲植被高光谱与浅层土壤含水量拟合研究[J].农业机械学报,2017,(9):1-16[CHEN Wenqian,DING Jianli,TAN Jiao,et al. Analysis vegetation of hyperspectral reflectance and shallow soil water content in arid area[J]. Transactions of the Chinese Society for Agricultural Machinery,2017,(9):1-16.]
[21]侯艳军,塔西甫拉提·特依拜,买买提·沙吾提,等.荒漠土壤有机质含量高光谱估算模型[J].农业工程学报,2014,30(16):113-120.[HOU Yanjun,TIYIP Tashpolat,SAWUT Mamat,et al. Estimation model of desert soil organic matter content using hyperspectral data[J]. Transactions of the Chinese Society of Agricultural Engineering,2014,30(16):113-120.]
[22]彭翔,胡丹,曾文治,等.基于EPO-PLS回归模型的盐渍化土壤含水率高光谱反演[J].农业工程学报,2016,32(11):167-173.[PENG Xiang,HU Dan,ZENG Wenzhi,et al. Estimating soil moisture from hyperspectra in saline soil based on EPO-PLS regression[J]. Transactions of the Chinese Society of Agricultural Engineering,2016,32(11):167-173.]
[23] ZENG W Z,HUANG J S,XU C,et al. Hyperspectral reflectance models for soil salt content by filtering methods and waveband selection[J]. Ecological Chemistry&Engineering S,2016,23(1):117-130.
[24] FARIFTEH J,MEER F V D,ATZBERGER C,et al. Quantitative analysis of salt-affected soil reflectance spectra:A comparison of two adaptive methods(PLSR and ANN)[J]. Remote Sensing of Environment,2007,110(1):59-78.
[2] ZHAO Z,TASHPOLAT T,ZHANG F,et al. Spectral characteristics of soil salt content in typical oasis of Tarim River’s middle reaches[J]. Journal of Natural Disasters,2012,21(5):72-78.
[3]吴明珠,李小梅,沙晋明.亚热带红壤全氮的高光谱响应和反演特征研究[J].光谱学与光谱分析,2013,33(11):3111-3115.[WU Mingzhu,LI Xiaomei,SHA Jinming. Spectral inversion models for prediction of red soil total nitrogen content in subtropical region(Fuzhou)[J]. Spectroscopy&Spectral Analysis,2013,33(11):3111-3115.]
[4] NAGY A,PETERRICZU,GALYA B,et al. Spectral estimation of soil water content in visible and near infra-red range[J]. 2014,3(3):163-171.
[5] ZHANG P,LI Y. Study on the comparisons of the establishment of two mathematical modeling methods for soil organic matter content based on spectral reflectance[J]. Spectroscopy&Spectral Analysis,2016,36(3):903-910.
[6]张晓光,黄标,季峻峰,等.基于可见近红外高光谱的东北盐渍土盐分定量模型研究[J].光谱学与光谱分析,2012,32(8):2075-2079.[ZHANG Xiaoguang,HUANG Biao,JI Junfeng,et al. Quantitative prediction of soil salinity content with visible-near infrared hyper-spectra in northeast China[J]. Spectroscopy&Spectral Analysis,2012,32(8):2075-2079.]
[7]杨爱霞,丁建丽,李艳红,等.基于表观电导率与实测光谱的干旱区湿地土壤盐分监测[J].中国沙漠,2016,36(5):1365-1373.[YANG Aixia,DING Jianli,LI Yanhong,et al. Apparent electronic conductivity and measured spectral for monitoring soil salt content in arid lakeside wetland[J]. Journal of Desert Research,2016,36(5):1365-1373.]
[8] GOLDSHLEGER N,CHUDNOVSKY A,BEN-BINYAMIN R. Predicting salinity in tomato using soil reflectance spectra[J]. International Journal of Remote Sensing,2013,34(17):6079-6093.
[9] NETO O R,TEIXEIRA A,LEAO R,et al. Hyperspectral remote sensing for detecting soil salinization using Pro SpecTIR-VS aerial imagery and sensor simulation[J]. Remote Sensing,2017,9(1):1-16.
[10]彭杰,王家强,向红英,等.土壤含盐量与电导率的高光谱反演精度对比研究[J].光谱学与光谱分析,2014,34(2):510-514.[PENG Jie,WANG Jiaqiang,XIANG Hongying,et al. Comparative study on hyperspectral inversion accuracy of soil salt content and electrical conductivity[J]. Spectroscopy and Spectral Analysis,2014,34(2):510-514.]
[11]屈永华,段小亮,高鸿永,等.内蒙古河套灌区土壤盐分光谱定量分析研究[J].光谱学与光谱分析,2009,29(5):1362-1366.[QU Yonghua,DUAN Xiaoliang,GAO Hongyong,et al.Quantitative retrieval of soil salinity using hyperspectral data in the region of Inner Mongolia Hetao Irrigation District[J]. Spectroscopy&Spectral Analysis,2009,29(5):1362-1366.]
[12]丁建丽,伍漫春,刘海霞,等.基于综合高光谱指数的区域土壤盐渍化监测研究[J].光谱学与光谱分析,2012,32(7):1918-1922.[DING Jianli,WU Manchun,LIU Haixia,et al. Study on the soil salinization monitoring based on synthetical hyperspectral index[J]. Spectroscopy&Spectral Analysis,2012,32(7):1918-1922.]
[13]段鹏程,熊黑钢,李荣荣,等.不同干扰程度的盐渍土与其光谱反射特征定量分析[J].光谱学与光谱分析,2017,37(2):571-576.[DUAN Pengcheng,XIONG Heigang,LI Rongrong,et al. A quantitative analysis of the reflectance of the saline soil under different disturbance extent[J]. Spectroscopy&Spectral Analysis,2017,37(2):571-576.]
[14]李亚莉,乔江飞,董天宇,等.不同质地盐渍化土壤水盐含量的高光谱反演[J].应用生态学报,2016,27(12):3807-3815.[LI Yali,QIAO Jiangfei,DONG Tianyu,et al. Hyperspectral inversion of soil water and salt content in soils with different textures[J]. Chinese Journal of Applied Ecology,2016,27(12):3807-3815.]
[15]刘娅,潘贤章,王昌昆,等.基于可见-近红外光谱的滨海盐土土壤盐分预测方法[J].土壤学报,2012,49(4):824-829.[LIU Ya,PAN Xianzhang,WANG Changkun,et al. Prediction of coastal saline soil salinity based on vis-nir reflectance spectroscopy[J]. Acta Pedologica Sinica,2012,49(4):824-829.]
[16] ZHAN X Y,ZHAO N,LIN Z Z,et al. Effect of algorithms for calibration set selection on quantitatively determining asiaticoside content in centella total glucosides by near infrared spectroscopy[J].Spectroscopy&Spectral Analysis,2014,34(12):3267-3272.
[17]尼珍,胡昌勤,冯芳.近红外光谱分析中光谱预处理方法的作用及其发展[J].药物分析杂志,2008,28(5):824-829.[NI Zhen,HU Changqin,FENG Fang. Progress and effect of spectral data pretreatment in NIR analytical technique[J]. Chinese Journal of Pharmaceutical Analysis,2008,28(5):824-829.]
[18]夏俊芳,李培武,李小昱,等.不同预处理对近红外光谱检测脐橙VC含量的影响[J].农业机械学报,2007,38(6):107-111.[XIA Junfang,LI Peiwu,LI Xiaoyu,et al. Effect of different pretreatment method of nondestructive measure vitamin C content of umbilical orange with near-infrared spectroscopy[J]. Transactions of the Chinese Society for Agricultural Machinery,2007,38(6):107-111.]
[19]刘娅,潘贤章,王昌昆,等.土壤湿润条件下基于光谱对称度的盐渍土盐分含量预测[J].光谱学与光谱分析,2013,33(10):2771-2776.[LIU Ya,PAN Xianzhang,WANG Changkun,et al.Predicting soil salinity based on spectral symmetry under wet soil condition[J]. Spectroscopy&Spectral Analysis,2013,33(10):2771-2776.]
[20]陈文倩,丁建丽,谭娇,等.干旱区绿洲植被高光谱与浅层土壤含水量拟合研究[J].农业机械学报,2017,(9):1-16[CHEN Wenqian,DING Jianli,TAN Jiao,et al. Analysis vegetation of hyperspectral reflectance and shallow soil water content in arid area[J]. Transactions of the Chinese Society for Agricultural Machinery,2017,(9):1-16.]
[21]侯艳军,塔西甫拉提·特依拜,买买提·沙吾提,等.荒漠土壤有机质含量高光谱估算模型[J].农业工程学报,2014,30(16):113-120.[HOU Yanjun,TIYIP Tashpolat,SAWUT Mamat,et al. Estimation model of desert soil organic matter content using hyperspectral data[J]. Transactions of the Chinese Society of Agricultural Engineering,2014,30(16):113-120.]
[22]彭翔,胡丹,曾文治,等.基于EPO-PLS回归模型的盐渍化土壤含水率高光谱反演[J].农业工程学报,2016,32(11):167-173.[PENG Xiang,HU Dan,ZENG Wenzhi,et al. Estimating soil moisture from hyperspectra in saline soil based on EPO-PLS regression[J]. Transactions of the Chinese Society of Agricultural Engineering,2016,32(11):167-173.]
[23] ZENG W Z,HUANG J S,XU C,et al. Hyperspectral reflectance models for soil salt content by filtering methods and waveband selection[J]. Ecological Chemistry&Engineering S,2016,23(1):117-130.
[24] FARIFTEH J,MEER F V D,ATZBERGER C,et al. Quantitative analysis of salt-affected soil reflectance spectra:A comparison of two adaptive methods(PLSR and ANN)[J]. Remote Sensing of Environment,2007,110(1):59-78.