Quantitative Analysis Method of Soil Elements Combining Sensitivity Dimensionality Reduction and Support Vector Regression

Fusheng Li; Xiaolong Zeng

doi:10.3788/LOP213241

Laser & Optoelectronics Progress, Volume. 60, Issue 5, 0530002(2023)

Quantitative Analysis Method of Soil Elements Combining Sensitivity Dimensionality Reduction and Support Vector Regression

Fusheng Li^1,2、* and Xiaolong Zeng^1,2

Author Affiliations

¹School of Automation Engineering, University of Electronic Science and Technology of China, Chengdu 611731, Sichuan, China

²Yangtze Delta Region Institute, University of Electronic Science and Technology of China, Huzhou 313099, Zhejiang, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(11)

Fig. 1. Spectra before and after background subtraction. (a) Original spectrum and estimated background; (b) original spectrum and corrected spectrum

Download full size

Fig. 2. Preprocessing effect after background subtraction

Download full size

Fig. 3. Flow chart of quantitative analysis method based on BOA-SVR

Download full size

Fig. 4. Physical image of the sample and XRF spectrometer. (a) Sample; (b) XRF spectrometer

Download full size

Fig. 5. Sensitivity analysis result of the As element

Download full size

Fig. 6. Prediction results of the model under different feature dimensions

Download full size

Fig. 7. Prediction results of Cu element. (a) SVR model with feature dimension reduction; (b) SVR model with all features as inputs; (c) PLS model

Download full size

Fig. 8. Prediction results of As element. (a) SVR model with feature dimension reduction; (b) SVR model with all features as inputs; (c) PLS model

Download full size

Table 1. Prediction results of three models on verification set in Cu element verification

View table

Table 1. Prediction results of three models on verification set in Cu element verification

No.	Reference value	Predictive value			Relative error
No.	Reference value	$S V R^{*}$	SVR	PLS	$S V R^{*}$	SVR	PLS
2	25.7	23.97	23.37	0.00	0.0672	0.0906	1.0000
5	916.0	932.84	908.24	1062.30	0.0184	0.0085	0.1597
7	7.2	9.84	66.27	0.00	0.3671	8.2047	1.0000
10	187.0	179.84	220.50	195.00	0.0383	0.1791	0.0428
12	54.2	54.88	74.23	60.07	0.0125	0.3696	0.1082
21	22.6	10.60	184.53	0.00	0.5308	7.1650	1.0000
24	118.0	112.03	133.18	108.47	0.0506	0.1286	0.0807
27	45.0	70.77	109.63	82.73	0.5727	1.4362	0.8384
32	177.0	170.59	131.64	184.14	0.0362	0.2563	0.0403
33	37.0	50.48	56.65	37.60	0.3642	0.5312	0.0162
48	29.0	29.44	29.54	39.15	0.0150	0.0187	0.3498
52	390.0	457.35	300.90	237.89	0.1727	0.2285	0.3900
53	11.4	19.69	12.83	0.00	0.7270	0.1259	1.0000

Table 2. Cu element prediction results obtained by three models
View table
Table 2. Cu element prediction results obtained by three models
Model $R_{C}^{M S E}$ $R_{C}^{2}$ $R_{P}^{M S E}$ $R_{P}^{2}$
$S V R^{*}$ 11.0334 0.9970 22.8803 0.9918
SVR 6.9356 0.9988 73.8296 0.9146
PLS 24.1319 0.9856 66.1133 0.9315

Table 3. As element prediction results obtained by three models
View table
Table 3. As element prediction results obtained by three models
Model $R_{C}^{M S E}$ $R_{C}^{2}$ $R_{P}^{M S E}$ $R_{P}^{2}$
$S V R^{*}$ 1.1271 0.9863 11.6868 0.9526
SVR 0.3038 0.9996 16.5271 0.7534
PLS 17.0948 0.4192 37.5909 0.4899

Tools

Get Citation

Copy Citation Text

Fusheng Li, Xiaolong Zeng. Quantitative Analysis Method of Soil Elements Combining Sensitivity Dimensionality Reduction and Support Vector Regression[J]. Laser & Optoelectronics Progress, 2023, 60(5): 0530002

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites