Estimate of soil heavy metal in a mining region using PCC-SVM-RFECV-AdaBoost combined with reflectance spectroscopy

Wang, Yueyue; Niu, Ruiqing; Lin, Guo; Xiao, Yingxu; Ma, Hangling; Zhao, Lingran

doi:10.1007/s10653-023-01488-w

Cited by 7 publications

(8 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The highest R 2 (0.884) was observed in the RLFD transformation. The smaller values of RMSE and MAE indicated that the estimation accuracy of the RFR model was high, and it had a high stability and generalization ability [33]. Overall, RFR based on RLFD (R 2 = 0.884, RMSE = 2.817%, MAE = 2.222) was the optimal estimation method for SOM content using the hyperspectral remote sensing data.…”

Section: Estimation Of Som Content With Rfrmentioning

confidence: 88%

See 1 more Smart Citation

Hyperspectral Estimation Model of Organic Matter Content in Farmland Soil in the Arid Zone

Subi,

Eziz,

Zhong

2023

Sustainability

View full text Add to dashboard Cite

Soil organic matter (SOM) is one of the most important indicators of soil quality. Hyperspectral remote sensing technology has been recognized as an effective method to rapidly estimate SOM content. In this study, 173 samples (0–20 cm) were collected from farmland soils in the northwestern arid zones of China. Partial least squares regression (PLSR), support vector machine regression (SVMR), and random forests regression (RFR), based on 15 types of mathematical transformations of the original spectral data of soil, were applied for identifying the optimal estimation method. Distribution of SOM content was mapped using both ground-measured values and predicted values estimated based on the optimum models. Obtained results indicated that the important spectral wavebands with the highest correlation were identified as 421 nm, 441 nm, 1014 nm, 1045 nm, and 2351 nm for SOM in the soil. Spectral transformations had obvious effects on the spectral characteristics of SOM. The optimal estimation was obtained when RFR was combined with the reciprocal logarithmic first-order differential (RLFD) (R2 = 0.884, RMSE = 2.817%, MAE = 2.222) for SOM contents. Finally, the RFR-RLFD method had much better performance compared with the PLSR and SVMR models. Results of this study can provide an alternative to the application of the hyperspectral estimation of SOM in farmland soils in arid zones.

show abstract

Section: Estimation Of Som Content With Rfrmentioning

confidence: 88%

“…RMSE and MAE were used to represent the predictive capacity and the robustness of the hyperspectral inversion models. In general, lower RMSE and MAE indicate better model prediction accuracy [33]. The analytical expressions of R 2 , RMSE, and MAE are given below:…”

Section: Algorithm Assessment Approachmentioning

confidence: 99%

Hyperspectral Estimation Model of Organic Matter Content in Farmland Soil in the Arid Zone

Subi,

Eziz,

Zhong

2023

Sustainability

View full text Add to dashboard Cite

show abstract

“…When RPD ≥ 2.0, the prediction model has a good prediction ability, when 1.4 ≤ RPD < 2.0, the model has the initial predictive capability, and when RPD < 1.4, the model has a poor predictive capability. In general, lower RMSE and MAE indicate better model prediction accuracy [26] .…”

Section: Model Validationmentioning

confidence: 97%

Hyperspectral Inversion and Analysis of Zinc Concentration in Urban Soil in the Urumqi City of China

Zhong,

Eziz,

Ainiwaer

et al. 2023

j. of environ. & earth. sci.

View full text Add to dashboard Cite

Excessive accumulation of zinc (Zn) in urban soil can lead to environmental pollution and pose a potential threat to human health and the ecosystem. How to quickly and accurately monitor the urban soil zinc content on a large scale in real time and dynamically is crucial. Hyperspectral remote sensing technology provides a new method for rapid and nondestructive soil property detection. The main goal of this study is to find an optimal combination of spectral transformation and a hyperspectral estimation model to predict the Zn content in urban soil. A total of 88 soil samples were collected to obtain the Zn contents and related hyperspectral data, and perform 18 transformations on the original spectral data. Then, select important wavelengths by Pearson's correlation coefficient analysis (PCC) and CARS. Finally, establish a partial least squares regression model (PLSR) and random forest regression model (RFR) with soil Zn content and important wavelengths. The results indicated that the average Zn content of the collected soil samples is 60.88 mg/kg. Pearson's correlation coefficient analysis (PCC) and CARS for the original and transformed wavelengths can effectively improve the correlations between the spectral data and soil Zn content. The number of important wavelengths selected by CARS is less than the important wavelengths selected by PCC. Partial least squares regression model based on first-order differentiation of the reciprocal by CARS (CARS-RTFD-PLSR) is more stable and has the highest prediction ability (R2 = 0.937, RMSE = 8.914, MAE = 2.735, RPD = 3.985). The CARS-RTFD-PLSR method can be used as a means of prediction of Zn content in soil in oasis cities. The results of the study can provide technical support for the hyperspectral estimation of the soil Zn content.

show abstract

“…Currently, these feature selection algorithms have been effectively applied in various fields, especially the RFECV algorithm has been widely used in several fields. In the geological sciences field [27], it has been used for feature selection in soil heavy metal contamination data. In the electric load field [28,29], it has been used to select.…”

mentioning

confidence: 99%

“…At present, among the three types of feature selection algorithms, Filter, Wrapper, and Embedded, the wrapped feature selection algorithm is widely used, especially the RFECV algorithm has become a hot application in the field of feature selection. In the field of geoscience, for the problem of soil heavy metal pollution, Yueyue Wang [27] et al used support vector machine recursive feature elimination cross-validation (SVM-RFECV) to select among pre-selected feature bands. In the field of electric load, Liang H [28] et al proposed a two-stage short-term load forecasting method based on the RFECV algorithm and time-convolutional network efficient channel attention mechanism-long and short-term memory network (TCN-ECA-LSTM).…”

mentioning

confidence: 99%

A More Flexible and Robust Feature Selection Algorithm

Tu,

Su,

Tang

et al. 2023

IEEE Access

View full text Add to dashboard Cite

With the rising number of actual data, the challenges of huge model operations and poor generalization capacity arise, making the selection of the right feature set a significant issue. This study proposes ImprovedRFECV, an enhanced feature selection approach for cross-validated recursive feature elimination (RFECV). The algorithm first increases the robustness of the optimal feature subset by randomly sampling different data, building multiple models, and comparing the scores. Simultaneously, L1 regularization term and L2 regularization term scores are introduced to thoroughly evaluate the value of each feature, reducing the influence on the anti-interference term and further improving the algorithm's accuracy and stability. Additionally, a multi-model ensemble learning framework is utilized to promote generalization ability and effectively prevent overfitting. Finally, a both-end expansion removal strategy is adopted to solve the problem of strong covariance of features while enhancing the flexibility of the algorithm. The experimental results show that, when compared to the RFECV algorithm, the ImprovedRFECV algorithm obtains fewer optimal average features and performs better on the optimal feature subset on five datasets from five different domains, demonstrating the algorithm's high robustness and generalization ability.

show abstract

Estimate of soil heavy metal in a mining region using PCC-SVM-RFECV-AdaBoost combined with reflectance spectroscopy

Cited by 7 publications

References 31 publications

Hyperspectral Estimation Model of Organic Matter Content in Farmland Soil in the Arid Zone

Hyperspectral Estimation Model of Organic Matter Content in Farmland Soil in the Arid Zone

Hyperspectral Inversion and Analysis of Zinc Concentration in Urban Soil in the Urumqi City of China

A More Flexible and Robust Feature Selection Algorithm

Contact Info

Product

Resources

About