Abstract:Soil composition is complex and varied. Predicting the contents of soil propertiesfast and efficiently is important for precision agriculture. Spectra are usually measured on dried soil samples. However, soil moisture is an important indicator for the guidance of agriculture activities. In order to predict the soil organic matter (SOM), soil moisture content (SMC), total iron (Fe) and pH value, we propose to measurement VIS-NIR spectra directly on wet samples and use Standard normal variable (SNV)-Continuous wavelet transform (CWT) method on spectra. CWT method uses Mexh as wavelet filter and 10 scales after SNV on each spectrum. Seven common methods, including Gauss filter (GS), First derivative (FD), Continuous removal (CR), and Mathematical transform (Log(1/R)) et al were used as comparisons. All of 74 samples were divided into 50 and 24, for calibrated and validation datasets. On the coefficients of each scale after SNV-CWT, wavebands that passed 0.05 significance level were selected as RF input variables. The optimal scale for each property was confirmed based on the statistical indicators of validation models. Then the Pearson correlation coefficients (PCC), Model based coefficients (MBC) and Grey relation degree (GRD) between each property and wavelet coefficients were calculated on the optimal scales. Models were estimated by the filter screening method based on the correlation coefficients calculated by the three methods. Results showed that, accuracies of all properties were improved after SNV-CWT comparing to the 7 commonly methods. The optimal transformation scales were 7, 8, 1 and 10, corresponding to SOM, SMC, Fe and pH respectively. When taking high dimension features as input variables, the Coefficient of Determination (R2) was reached to 0.90 and 0.93. The best analysis method was MBC. Because the models performed best when wavebands for the models were selected using MBC as a screening method, the R2 of SOM and SMC was 0.94 and the accuracies of Fe (R2=0.67, Mse=0.01%, RPD=1.76) and pH (R2=0.80, Mse=0.1, RPD=2.24) were greatly improved, methods can be used for extracting and monitoring multi soil properties.
Key words:Black soil area; Vis-Nir spectra; Data processing; Soil organic matter; Soil moisture; Total iron; pH
谭 洋,姜琦刚,刘骅欣,刘 斌,高 鑫,张 博. 基于多尺度SNV-CWT特征的黑土有机质、水分、总铁及pH值估测[J]. 光谱学与光谱分析, 2021, 41(11): 3424-3430.
TAN Yang, JIANG Qi-gang, LIU Hua-xin, LIU Bin, GAO Xin, ZHANG Bo. Estimation of Organic Matter, Moisture, Total Iron and pH From Back Soil Based on Multi Scales SNV-CWT Transformation. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(11): 3424-3430.
[1] Ben-Dor E, Patkin K, Banin A, et al. International Journal of Remote Sensing, 2002, 23(6): 1043.
[2] Morellos A, PantaziX E, Moshou D, et al. Biosystems Engineering, 2016, 152: 104.
[3] Gholizadeh A, Luboš B, Mohammadmehdi S, et al. Applied Spectroscopy,2013, 67(12): 1349.
[4] Stenberg B, Raphael A, Viscarra R, et al. In Advances in Agronomy, 2010, 107: 163.
[5] Wang Y, Huang T, Liu J, et al. Computers and Electronics in Agriculture, 2015, 111: 69.
[6] Kahaer Y, Tashpolat N, Shi Q, et al. Water, 2020, 12(12): 3360.
[7] Wang T, Jiao L, Yan C, et al. Chemometrics and Intelligent Laboratory Systems, 2019, 194: 103854.
[8] De Santana, De Giuseppe, de Souza, et al. Microchemical Journal, 2018, 145: 1094.
[9] Vašát, Radim, Radka Kodešová, et al. Geoderma 2017, 298: 46.
[10] Rivard B, Feng J, Gallie A, et al. Remote Sensing of Environment,2008, 112(6): 2850.
[11] Cheng T, Rivard B, Sánchez G A, et al. Remote Sensing of Environment, 2010, 114(4): 899.
[12] Blackburn G A, Ferwerda J G. Remote Sensing of Environment, 2008, 112(4): 1614.
[13] Tom Fearn, Cecilia Riccioli, Ana Garrido-Varo, et al, Chemometrics and Intelligent Laboratory Systems, 2009, 96(1):22.
[14] CHEN Yi-yun, QI Tian-ci, HUANG Ying-jing, et al(陈奕云, 齐天赐, 黄颖菁, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报),2017, 33(6): 107.