Carbonate content in soil is an important basis for soil classification and fertility evaluation. Based on an analysis of calcium carbonate content, the authors chose 78 soil samples from Loess Plateau of Shaanxi Province as the research objects. The visible near infrared hyperspectral reflectance (350~2 500 nm) data of soil samples were obtained by hyperspectral imager. Three mathematical transformations, i.e., first-order differentiation, second-order differentiation and continuum removal, were carried out on the original spectral curve, and correlation analysis was used. The method and the continuous projection algorithm were used to select the sensitive band respectively, and the Stochastic Forest regression was used to establish the estimation model of soil calcium carbonate. According to the results obtained, the spectral curve characteristics of Huangmian soil are almost the same, there are obvious absorption characteristics at 1 440 nm, 1 900 nm, 2 200 nm and so on, and the calcium carbonate content and spectral reflectance show a positive correlation trend; the accuracy of random forest estimation model based on the second-order differential and continuous projection algorithm is the highest, the validation set R2 is 0.82, and the PRD value is 2.37.
Chen L M, Zhang G L. Determination of parent material homogeneity and evolution of characteristics of paddy soils developed in coastal sediments[J]. Soil Journal, 2009,46(5):753-763.
Zhang N, He X D, Wu W. Characteristics of organic matter and calcium carbonate in three soils of Tengger Desert[J]. Acta Ecologica, 2009,29(8):4095-4101.
Tong Q X, Zhang B, Zhang L F. Current progress of hyperspectral remote sensing in China[J]. Journal of Remote Sensing, 2016,20(5):689-707.
[5]
Gaffey S J. Spectral reflectance of carbonate minerals in visible and near infrared(0.35~2.55 μm):A nhydrous carbonate minerals[J]. Journal of Geophysical Research, 1987,92:1429— 1440.
[6]
张婷婷. 基于PLS模型的农业土壤成分高光谱遥感反演研究[D]. 长春:吉林大学, 2010.
Zhang T T. Partial least squares modeling of hyperspectral remote sensing for mapping agricultural soil properties[D]. Changchun:Jilin University, 2010.
[7]
Stenberg B, Viscarra Rossel R A, Mouazen A M, et al. Visible and near infrared spectroscopy in soil science[J]. Advances in A gronomy, 2010,107:163— 215.
[8]
Sexton J, Laake P. Standard errors for bagged and random forest estimators[J]. Computational Statistics & Data Analysis, 2009,53(3):801-811.
Song J G, Gao Q S, Li Z. Application of stochastic forest regression in seismic reservoir prediction[J]. Petroleum Geophysical Exploration, 2016,51(6):1203-1211.
Qin Q R. Prediction of soil free iron and calcium carbonate content based on hyperspectral data in Shaanxi Province[D]. Yang Ling:Northwest Agricultural and Forestry University of Science and Technology, 2019.
Shan H B, Jiang P A, Yan A, et al. Hyperspectral retrieval of organic matter in grey desert soil of oasis farmland in Northern Xinjiang[J]. Journal of Agricultural Resources and Environment, 2018,35(3):276-282.
[12]
Araújo M C U, Saldanha T C B, Galvão R K H, et al. The successive projections algorithm for variable selection in spectroscopic multicomponent analysis[J]. Chemometrics and Intelligent Laboratory Systems, 2001,57(2):65-73.
Yang H Y, Zhou Q, Yang J, et al. Hyperspectral diagnosis of nitrogen nutrition in rice leaves[J]. Zhejiang Journal of Agriculture, 2019,31(10):1575-1582.
Li X L, Yi S L, He S L, et al. Identification of pummelo cultivars based on hyperspectral imaging technology[J]. Spectroscopy and Spectral Analysis, 2015,35(9):2639-2643.
pmid: 26669182
Hong C Q, Zheng G, Chen C C. Study on reflectance spectrum estimation of calcium carbonate content in coastal soil of Northern Jiangsu[J]. Journal of Soil Science, 2016,53(5):1121-1129.
Zhang D, Zhang F, et al. Application of fractional differential in preprocessing hyperspec-tral data of saline soil[J]. Transactions of the Chinese Society of Agricultural Engineering, 2014,30(24):151-160.
[19]
Gomez C, Philippe L, Guillaume C. Continuum removal versus PLSR method for clay and calcium carbonate content estimation from laboratory and airborne hyperspectral measurements[J]. Geoderma, 2008,148(2):141-148.