Prediction of lead content in soil based on model population analysis coupled with ELM algorithm
XIAO Yehui1(), SONG Nidi2, MENG Panpan2, WANG Peijun2, FAN Shenglong2()
1. College of Resource and Environment, Fujian Agriculture and Forestry University, Fuzhou 350007, China 2. College of Public Management, Fujian Agriculture and Forestry University, Fuzhou 350007, China
This paper aims to explore the optimal inversion model of regional heavy metal content in soil. With Longhai City taken as the study area, this study preprocessed the original spectral data of soil using the methods of Savizky Golay (SG), wavelet transform (WT), gaussian filter (GF), and multiple scatter correction (MSC) individually, then eliminated the interference and wavelength bearing no information using the wavelength selection algorithms developed based on model population analysis (MPA), including the competitive adaptive reweighted sampling (CARS), variable iterative space shrinkage approach (VISSA), iteratively variable subset optimization (IVSO), and interval combination optimization (ICO), and finally predicted the lead content in soil using the linear partial least squares regression (PLSR) model, nonlinear support vector machine (SVM) model, and extreme learning machine (ELM) based on neural network. The results are as follows. ① Among the inversion models of lead content in soil established using various preprocessing methods, the model built based on reconstructed spectral data of level 7th by wavelet transform had the most optimal prediction accuracy, with R2=0.736, RMSE=5.426, RPD=1.976, and RPIQ=2.560. ② The CARS, VISSA, IVSO, and ICO algorithms developed based on MPA significantly improved the performance of model interpretation and generalization and improved modeling efficiency. ③ In terms of overall prediction results, the three regression models were in the order of ELM>PLSR>SVM. Among them, the ICO-ELM had the highest prediction accuracy, with R2=0.863, RMSE=3.953, RPD=2.712,and RPIQ=3.514. Therefore, the optimal model established in this study can provide a new theoretical reference for the rapid monitoring of regional land quality and ecological indicators.
肖烨辉, 宋妮迪, 孟盼盼, 王培俊, 范胜龙. 模型集群分析策略联合ELM的土壤重金属Pb含量预测[J]. 自然资源遥感, 2021, 33(4): 143-152.
XIAO Yehui, SONG Nidi, MENG Panpan, WANG Peijun, FAN Shenglong. Prediction of lead content in soil based on model population analysis coupled with ELM algorithm. Remote Sensing for Natural Resources, 2021, 33(4): 143-152.
He J L, Han C S, Wei R, et al. Research on indirect hyperspectral estimating model of heavy metal Cd based on partial least squares regression[J]. Remote Sensing for Land and Resources, 2019,31(4):96-103.doi: 10.6046/gtzyyg.2019.04.13.
doi: 10.6046/gtzyyg.2019.04.13
Gao K X, Jiao H M, Wang X C. Inversion model of forest canopy height based on image texture,spectral and topographic features[J]. Remote Sensing for Land and Resources, 2020,32(3):63-70.doi: 10.6046/gtzyyg.2020.03.09.
doi: 10.6046/gtzyyg.2020.03.09
Duan H W, Zhu R G, Xu W D, et al. Hyperspectral imaging detection of total viable count from vacuum packing cooling mutton based on GA and CARS algorithms[J]. Spectroscopy and Spectral Analysis, 2017,37(3):847-852.
[5]
Le B T, 肖冬, 毛亚纯, 等. 可见、近红外光谱和深度学习CNN-ELM算法的煤炭分类[J]. 光谱学与光谱分析, 2018,38(7):2107-2112.
Le B T, Xiao D, Mao Y C, et al. Coal classification based on visible,near-infrared spectroscopy and CNN-ELM algorithm[J]. Spectroscopy and Spectral Analysis, 2018,38(7):2107-2112.
Wang L S, Lu C P, Wang R J, et al. Optimization for Vis/NIRS prediction model of soil available nitrogen content[J]. Chinese Journal of Luminescence, 2018,39(7):1016-1023.
doi: 10.3788/fgxb
Wu Q, Jiang Q G, Shi P F, et al. Estimation of soil calcium carbonate content based on hyperspectral data[J]. Remote Sensing for Land and Resources, 2021,33(1):138-144.doi: 10.6046/gtzyyg.2020095.
doi: 10.6046/gtzyyg.2020095
Li P, Zhou J, Jiang L W, et al. A variable selection approach of near infrared spectra based on window competitive adaptive reweighted sampling strategy[J]. Spectroscopy and Spectral Analysis, 2019,39(5):1428-1432.
[9]
Li H, Liang Y, Xu Q, et al. Model population analysis for variable selection[J]. Journal of Chemometrics, 2010,24(7-8):418-423.
doi: 10.1002/cem.1300
Yu L, Zhang T, Zhu Y X, et al. Determination of soybean leaf SPAD value using characteristic wavelength variables preferably selected by IRIV algorithm[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018,34(16):148-154.
Liu G S, Zhang C, Fan N Y, et al. Hyperspectral model optimization for tenderness of chilled tan-sheep mutton based on IVISSA[J]. Spectroscopy and Spectral Analysis, 2020,40(8):2558-2563.
Yun Y H, Deng B C, Liang Y Z. Progress of chemical modeling and model population analysis[J]. Chinese Journal of Analytical Chemistry, 2015,43(11):1638-1647.
[13]
Deng B C, Yun Y H, Liang Y Z, et al. A novel variable selection approach that iteratively optimizes variable space using weighted binary matrix sampling[J]. Analyst, 2014,139(19):4836-4845.
doi: 10.1039/C4AN00730A
[14]
Wang W, Yun Y, Deng B, et al. Iteratively variable subset optimization for multivariate calibration[J]. RSC Advances, 2015,5(116):95771-95780.
doi: 10.1039/C5RA08455E
[15]
Song X Z, Yue H, Hong Y, et al. A novel algorithm for spectral interval combination optimization[J]. Analytica Chimica Acta, 2016,948:19-29.
doi: 10.1016/j.aca.2016.10.041
Sun K, Sun B B, Zhou G H, et al. Study on concentration characteristics and influencing factors of heavy metals in soils in Longhai,Fujian Province[J]. Geoscience, 2018,32(6):1302-1310.
Liu Z C, Cai W S, Shao X G. Identification of NIR outlier samples by MCCV[J]. Science in China Series B:Chemistry, 2008,38(4):316-323.
[18]
Huang G, Huang G, Song S, et al. Trends in extreme learning machines:A review[J]. Neural Networks, 2015,61:32-48.
pmid: 25462632
[19]
Tan K, Wang H, Zhang Q, et al. An improved estimation model for soil heavy metal(loid) concentration retrieval in mining areas using reflectance spectroscopy[J]. Journal of Soils and Sediments, 2018,18(5):2008-2022.
doi: 10.1007/s11368-018-1930-6
Song X Z, Tang G, Zhang L D, et al. Research advance of variable selection algorithms in near infrared spectroscopy analysis[J]. Spectroscopy and Spectral Analysis, 2017,37(4):1048-1052.