# Hyperspectral Detection of Moisture Content in Rice Straw Nutrient Bowl Trays Based on PSO-SVR

^{1}

^{2}

^{*}

## Abstract

**:**

^{2}) and mean square error (RMSE) as evaluation indicators. The research results indicate that the PCA-PSO-SVR model established based on SG method preprocessing has the best predictive performance, with a training set decision coefficient R

^{2}C of 0.984, a training set mean square error RMSE-C of 2.775, a testing set decision coefficient R

^{2}P of 0.971, and a testing set mean square error RMSE-P of 3.448. The model therefore has a high accuracy. This study achieved rapid detection of water content in rice straw nutrition trays. This method provides a reliable theoretical basis and technical support for the rapid detection of rice straw nutrient bowl tray moisture content, and is of great significance for improving the quality of rice straw nutrient bowl trays; promoting the popularization and application of raising rice straw nutrient bowl tray seedlings and whole process mechanized planting technology system; improving soil structure; and protecting the ecological environment.

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Sample Preparation

#### 2.2. Determination of Moisture Content in Rice Seedling Trays

_{c}is the moisture content of the material on a wet basis at time t, %; m

_{t}is the mass of the material at time t, g; and m

_{d}is the dry matter mass of the material, g.

#### 2.3. Spectral Data Collection and Region of Interest Selection

#### 2.4. Hyperspectral Preprocessing Methods

#### 2.5. Dimension Reduction Method

#### 2.6. Modeling Methods

#### 2.6.1. Random Forest Regression

_{1}and D

_{2}on both sides. We found the feature and feature value partition point corresponding to the minimum sum of the mean squared deviations of D

_{1}and D

_{2}, and the expression is:

_{1}is the sample output mean of the D

_{1}dataset and c

_{2}is the sample output mean of the D

_{2}dataset.

_{1}and D

_{2}subspaces until there were no features to partition; we constructed a regression tree with the nodes at this time.

#### 2.6.2. Particle Swarm Optimization Support Vector Regression Model

_{1}and c

_{2}are acceleration constants used to adjust the maximum step size of learning; r

_{1}and r

_{2}are random functions used to increase search randomness; and $w\text{}$is the inertia weight, with a non-negative value, used to adjust the search range of the solution space.

#### 2.6.3. Xgboost Model

#### 2.6.4. Establishment and Evaluation of Regression Models

^{2}) and root mean square error (RMSE) as evaluation indicators to analyze the fitting effect of the regression model, and then determined the quality of the model.

_{i}represents the measured value of the i-th sample; y represents the predicted value of the i-th sample; $\overline{y}$ represents the average measurement value; and n represents the number of samples.

## 3. Results

#### 3.1. Data Statistics and Analysis

#### 3.2. Spectral Preprocessing

#### 3.3. Spectral Data Dimensionality Reduction

#### 3.4. Detection Results Based on Different Spectral Preprocessing Methods and Dimensionality Reduction Methods Modeling

## 4. Discussion

#### 4.1. Best Model Analysis

^{2}C) and the test set determination coefficient (R

^{2}P) is smaller when the pre-treated spectral data modeling is used. The RF and PSO-SVR models established using spectral data preprocessed by SG have better performance, while the XGBoost model established using unprocessed spectral data has better performance. The R

^{2}P of the model established using dimensionality reduced data is larger, indicating that the model established using dimensionality reduced data has a better predictive performance. The model accuracy of spectral data modeling extracted by the PCA algorithm is generally higher than that of spectral data modeling processed by CARS, which may be due to the fact that the feature matrix extracted by PCA contains more features related to the moisture content of rice seedling trays.

^{2}P increases by 0.085 and RMSE-P decreases by 3.384.

#### 4.2. Limitations and Future Work

## 5. Conclusions

^{2}P of all models can reach above 0.88, indicating that hyperspectral imaging technology can be used for detecting the moisture content of rice seedling trays. (2) The models R

^{2}C and R

^{2}P established using raw spectral data have a significant difference, while the model established using preprocessed spectral data has a smaller difference between R

^{2}C and R

^{2}P. Within the experimental scope of this study, SG preprocessing is more suitable for the establishment of RF and PSO-SVR models, and unprocessed spectral data are more suitable for the establishment of XGBoost models. (3) Overall, the model established using dimensionality reduced spectral data has a larger coefficient of determination and better performance. Among them, the model established using PCA-extracted feature variables has the best effect, indicating that within the scope of this experiment, PCA is more suitable for feature extraction of rice seedling tray moisture content compared to CARS. (4) The optimal moisture content detection model determined in this study is SG-PCA-PSO-SVR (R

^{2}C = 0.984, RMSE-C = 2.775, R

^{2}P = 0.971, RMSE-P = 3.448), which has the best predictive effect.

## Author Contributions

## Funding

## Conflicts of Interest

## Abbreviation

MSC | multivariate scattering correction |

SNV | standardization normal variables |

SG | Savitzky–Golay convolution smoothing |

PCA | principal component analysis |

CARS | competitive adaptive reweighting |

RF | random forest regression |

PSO-SVR | particle swarm optimization support vector regression |

## Appendix A

Preprocessing Method | Number of Feature Bands | n Estimators | Max Depth | Min Samples Leaf | Min Samples Split | Max Features | Max Leaf Nodes |
---|---|---|---|---|---|---|---|

None | No | 116 | 16 | 1 | 2 | 48 | 14 |

PCA | 50 | 9 | 1 | 2 | 10 | 49 | |

CARS | 80 | 10 | 1 | 2 | 9 | 44 | |

MSC | None | 185 | 6 | 1 | 4 | 38 | 42 |

PCA | 118 | 10 | 1 | 2 | 7 | 39 | |

CARS | 50 | 8 | 1 | 2 | 49 | 39 | |

SNV | None | 145 | 6 | 1 | 4 | 38 | 42 |

PCA | 43 | 15 | 1 | 2 | 6 | 48 | |

CARS | 22 | 10 | 1 | 2 | 34 | 49 | |

SG | None | 55 | 13 | 1 | 2 | 48 | 41 |

PCA | 51 | 10 | 1 | 2 | 10 | 49 | |

CARS | 87 | 7 | 1 | 2 | 48 | 34 |

Preprocessing Method | Number of Feature Bands | C | γ | ε |
---|---|---|---|---|

NO | NO | 10 | 4.870 | 0.209 |

PCA | 10 | 0.100 | 0.209 | |

CARS | 10 | 0.382 | 0.540 | |

MSC | NO | 10 | 0.820 | 0.123 |

PCA | 10 | 0.408 | 0.143 | |

CARS | 10 | 2.924 | 1 | |

SNV | NO | 10 | 1.054 | 0.177 |

PCA | 10 | 1.044 | 0.184 | |

CARS | 10 | 2.129 | 1 | |

SG | NO | 10 | 0.562 | 0.209 |

PCA | 10 | 0.2 | 0.209 | |

CARS | 10 | 0.416 | 0.619 |

Preprocessing Method | Number of Feature Bands | Learning_Rate | Subsample | Max_Depth | n_Estimators |
---|---|---|---|---|---|

None | None | 0.1 | 0.3 | 7 | 50 |

PCA | 0.1 | 0.5 | 9 | 50 | |

CARS | 0.1 | 0.5 | 11 | 50 | |

MSC | None | 1 | 0.4 | 7 | 45 |

PCA | 0.1 | 0.4 | 19 | 50 | |

CARS | 1 | 0.5 | 5 | 50 | |

SNV | None | 0.1 | 0.3 | 7 | 50 |

PCA | 0.1 | 0.3 | 7 | 50 | |

CARS | 0.1 | 0.4 | 9 | 50 | |

SG | None | 0.1 | 0.3 | 7 | 50 |

PCA | 0.1 | 0.2 | 5 | 50 | |

CARS | 0.1 | 0.5 | 13 | 50 |

## Appendix B

## References

- Qian, Y.T.; Zhang, Y.P.; Du, J.; Kong, X.P.; Chang, Z.P.; Ye, X.M. Advantages and disadvantages analysis of comprehensive utilization of straw in Jiangsu Province and countermeasure suggestions for collection-storage-transportation system. Trans. Chin. Soc. Agric. Eng.
**2019**, 35, 154–160. [Google Scholar] - Wang, J.W.; Tang, H.; Wang, J.F. Comprehensive Utilization Status and Development Analysis of Crop Straw Resource in Northeast China. Trans. Chin. Soc. Agric. Mach.
**2017**, 48, 1–21. [Google Scholar] - Goodman, B.A. Utilization of Waste Straw and Husks from Rice Production: A Review. J. Bioresour. Bioprod.
**2020**, 5, 143–162. [Google Scholar] [CrossRef] - Yu, H.M.; Li, H.Y.; Zhang, X.Y.; Wang, H.Y.; Sun, Y. Drying Kinetic Model of Microwave Coupled with Hot Air Drying of Straw-based Nutrient Seedling-growing Bowl Tray. Trans. Chin. Soc. Agric. Mach.
**2020**, 51, 339–348. [Google Scholar] - Zhang, Y.; Zhu, Q.J.; Gao, S.; Liu, S.; Li, L.H.; Chen, H.T. Optimization of Technological Parameters of Straw Fiber-Based Plant Fiber Seedling Pot Raw Materials. Appl. Sci.
**2021**, 11, 7152. [Google Scholar] [CrossRef] - Yu, H.M.; Li, H.Y.; Wang, C.; Li, H.L.; Zhang, X.Y.; Liang, Q.; Yu, B.B. Simulation Analysis of Flow Field Uniformity in Air Distribution Room of Rice Seedbed Microwave Hot Air Coupling Dryer. J. Agric. Mech. Res.
**2020**, 42, 15–21. [Google Scholar] - Yu, H.M.; Wang, C.; Han, Z.X.; Sun, Y.; Zhang, W.; Hu, J.; Liu, T.X. Optimization of steam drying conditions for seedling-growing tray made of paddy-straw. Trans. Chin. Soc. Agric. Eng.
**2013**, 29, 40–49. [Google Scholar] - Sun, H.; Liu, N.; Wu, L.; Zheng, T.; Li, M.Z.; Wu, J.Z. Visualization of Water Content Distribution in Potato Leaves Based on Hyperspectral Image. Spectrosc. Spectr. Anal.
**2019**, 39, 910–916. [Google Scholar] - Wei, Y.Z.; Wu, F.Y.; Xu, J.; Sha, J.; Zhao, Z.; He, Y.; Li, X. Visual Detection of the Moisture Content of Tea Leaves with Hyperspectral Imaging Technology. J. Food Eng.
**2019**, 248, 89–96. [Google Scholar] [CrossRef] - Zhang, J.; Tian, H.Q.; Zhao, Z.Y.; Zhang, L.N.; Zhang, J.; Li, F. Moisture content dectection in silage maize raw material based on hyperspectrum and improved discrete particle swarm. Trans. Chin. Soc. Agric. Eng.
**2019**, 35, 285–293. [Google Scholar] - Dong, C.; An, T.; Yang, M.; Yang, C.; Liu, Z.; Li, Y.; Duan, D.; Fan, S. Quantitative Prediction and Visual Detection of the Moisture Content of Withering Leaves in Black Tea (Camellia Sinensis) with Hyperspectral Image. Infrared Phys. Technol.
**2022**, 123, 104118. [Google Scholar] [CrossRef] - Li, H.; Zhang, K.; Chen, C.; Zhang, Z.Z.; Liu, Z.P. Detection of Moisture Content in Lettuce Canopy Based on Hyperspectral lmaging Technique. Trans. Chin. Soc. Agric. Mach.
**2021**, 52, 211–217. [Google Scholar] - Zhang, Z.T.; Wang, H.F.; Arnon, K.; Chen, J.Y.; Han, W.T. Inversion of Soil Moisture Content from Hyperspectra Based on Ridge Regression. Trans. Chin. Soc. Agric. Eng.
**2018**, 49, 240–248. [Google Scholar] - Liu, D.; Qu, J.; Sun, D.W.; Pu, H.; Zeng, X.A. Non-Destructive Prediction of Salt Contents and Water Activity of Porcine Meat Slices by Hyperspectral Imaging in a Salting Process. Innov. Food Sci. Emerg. Technol.
**2013**, 20, 316–323. [Google Scholar] [CrossRef] - Wu, T.; Yu, J.; Lu, J.; Zou, X.; Zhang, W. Research on Inversion Model of Cultivated Soil Moisture Content Based on Hyperspectral Imaging Analysis. Agriculture
**2020**, 10, 292. [Google Scholar] [CrossRef] - Wu, D.; Shi, H.; Wang, S.; He, Y.; Bao, Y.; Liu, K. Rapid Prediction of Moisture Content of Dehydrated Prawns Using Online Hyperspectral Imaging System. Anal. Chim. Acta
**2012**, 726, 57–66. [Google Scholar] [CrossRef] [PubMed] - Huang, M.; Wang, Q.; Zhang, M.; Zhu, Q. Prediction of Color and Moisture Content for Vegetable Soybean during Drying Using Hyperspectral Imaging Technology. J. Food Eng.
**2014**, 128, 24–30. [Google Scholar] [CrossRef] - Raponi, F.; Moscetti, R.; Monarca, D.; Colantoni, A.; Massantini, R. Monitoring and Optimization of the Process of Drying Fruits and Vegetables Using Computer Vision: A Review. Sustainability
**2017**, 9, 2009. [Google Scholar] [CrossRef] - Qian, W.; Yang, S.K.; Liu, G.L.; He, G.; Wang, Z.H.; Du, J.Q. Optimization of Drying Process Parameters of Solar Herbage Dry Equipment. Trans. Chin. Soc. Agric. Mach.
**2018**, 49, 110–118. [Google Scholar] - Tian, X.Y.; Huang, X.Y.; Bai, J.W.; Lv, R.Q.; Sun, Z.Y. Detection of Anthocyanin Content of Purple Sweet Potato during Storage Period Based on Near Infrared Spectroscopy. Trans. Chin. Soc. Agric. Mach.
**2019**, 50, 350–355. [Google Scholar] - Bai, T.C.; Yu, C.L.; Wang, X.P.; Wang, Y.M.; Zhang, N.N.; Yao, N. Near Infrared Spectrum Detection Method for Moisture Content of Populus Euphratica Leaf. Spectrosc. Spectr. Anal.
**2017**, 37, 3419–3423. [Google Scholar] - Fearn, T.; Riccioli, C.; Garrido-Varo, A.; Guerrero-Ginel, J.E. On the Geometry of SNV and MSC. Chemom. Intell. Lab. Syst.
**2009**, 96, 22–26. [Google Scholar] [CrossRef] - Hui, G.Y.; Sun, L.J.; Wang, J.N.; Wang, L.K.; Dai, C.J. Research on the Pre-Processing Methods of Wheat Hardness Prediction Model Based on Visible-Near Infrared Spectroscopy. Spectrosc. Spectr. Anal.
**2016**, 36, 2111–2116. [Google Scholar] - Chen, J.Y.; Xing, Z.; Zhang, Z.T.; Lao, C.C.; Li, X.W.; Wang, H.F. Comprehensive Evaluation of Waste Water Quality Based on Quantitative Inversion Model Hyperspectral Technology. Trans. Chin. Soc. Agric. Mach.
**2019**, 50, 200–209. [Google Scholar] - Li, Y.; Tan, X.; Zhang, W.; Jiao, Q.; Xu, Y.; Li, H.; Zou, Y.; Yang, L.; Fang, Y. Research and Application of Several Key Techniques in Hyperspectral Image Preprocessing. Front. Plant Sci.
**2021**, 12, 627865. [Google Scholar] [CrossRef] [PubMed] - Uddin, M.P.; Mamun, M.A.; Hossain, M.A. PCA-Based Feature Reduction for Hyperspectral Remote Sensing Image Classification. IETE Tech. Rev.
**2021**, 38, 377–396. [Google Scholar] [CrossRef] - Yan, S.A.; Shi, M.Z.; Lin, X.X.; Li, W.; Yao, Q.H. Principal Component Analysis and Cluster Analysis for Evaluating Amino Acid of Different Table Grapes (Vitis vinifera L.) Varieties. Sci. Technol. Food Ind.
**2022**, 43, 372–379. [Google Scholar] - Napoli, C.D.; Pope, I.; Masia, F.; Watson, P.; Langbein, W.; Borri, P. Hyperspectral and Differential CARS Microscopy for Quantitative Chemical Imaging in Human Adipocytes. Biomed. Opt. Express
**2014**, 5, 1378–1390. [Google Scholar] [CrossRef] - Liang, L.; Wei, L.; Fang, G.; Xu, F.; Deng, Y.; Shen, K.; Tian, Q.; Wu, T.; Zhu, B. Prediction of Holocellulose and Lignin Content of Pulp Wood Feedstock Using near Infrared Spectroscopy and Variable Selection. Spectrochim. Acta Part A Mol. Biomol. Spectrosc.
**2020**, 225, 117515. [Google Scholar] [CrossRef] - Wang, T.J.; Fang, K.; Yang, Y.; Zhang, X.D. Multi-spectral remote sensing inversion of soil heavy metal content using random forest regression model. Bull. Surv. Mapp.
**2021**, 92–95. [Google Scholar] [CrossRef] - Shourian, M.; Mousavi, S.J. Performance Assessment of a Coupled Particle Swarm Optimization and Network Flow Programming Model for Optimum Water Allocation. Water Resour Manag.
**2017**, 31, 4835–4853. [Google Scholar] [CrossRef] - Bayrami, B. Estimation of splitting tensile strength of modified recycled aggregate concrete using hybrid algorithms. Available SSRN 3992623
**2022**, 44, 375–392. [Google Scholar] [CrossRef] - Shi, X.; Yu, X.; Esmaeili-Falak, M. Improved Arithmetic Optimization Algorithm and Its Application to Carbon Fiber Reinforced Polymer-Steel Bond Strength Estimation. Compos. Struct.
**2023**, 306, 116599. [Google Scholar] [CrossRef] - Chen, P.; Qi, C.; Liu, R.W.; Wang, Z.Z.; Luo, H.; Yuan, J.J.; Liu, J.P. Quantitative Analysis of Carbon Content in Fly Ash Using LIBS Based on Support Vector Machine Regression. Acta Opt. Sin.
**2022**, 42, 278–285. [Google Scholar] - Qian, L.S.; Jiang, H.; Chang, S.S.; Li, D.; Wang, C.Y.; Chen, J.Y.; Dai, X.M. Extracting field-scale crop distribution in Lingnan using spatiotemporal filtering of Sentinel-1 time-series data. Trans. Chin. Soc. Agric. Eng.
**2022**, 38, 158–166. [Google Scholar] - Esmaeili-Falak, M.; Benemaran, R.S. Ensemble Deep Learning-Based Models to Predict the Resilient Modulus of Modified Base Materials Subjected to Wet-Dry Cycles. Geomech. Eng.
**2023**, 32, 583–600. [Google Scholar]

**Figure 2.**(

**a**) Hyperspectral imaging of Rice Straw Nutrient Bowl Tray; (

**b**) schematic diagram of ROI selection for rice seedling tray samples.

Preprocessing Method | Number of Feature Bands | RMSECV_min | Best Sampling Times | Feature Bands (nm) |
---|---|---|---|---|

NO | 50 | 2.8737 | 83 | 472 498 504 505 514 546 567 573 579 588 589 590 592 594 596 597 598 604 609 610 611 612 618 619 620 621 639 640 650 651 655 661 662 665 678 684 685 693 694 695 728 742 743 744 749 792 839 850 940 957 |

MSC | 45 | 3.4966 | 87 | 465 466 486 488 490 497 501 503 504 516 523 554 574 575 581 592 595 596 598 607 611 617 619 620 622 665 684 691 693 696 701 702 703 706 729 731 743 747 817 818 821 825 848 850 939 |

SNV | 82 | 3.4827 | 65 | 463 488 490 494 495 497 501 502 503 504 506 516 528 532 537 542 547 552 554 556 560 572 578 579 581 595 598 607 611 617 618 619 620 622 634 639 640 641 650 651 653 655 657 659 660 661 682 684 689 690 691 692 693 694 695 696 699 701 702 728 729 730 731 743 783 788 789 798 816 817 821 822 825 842 846 847 848 850 876 951 952 957 |

SG | 60 | 2.8522 | 76 | 472 476 480 485 488 495 498 503 505 517 531 567 568 573 580 586 592 596 597 598 599 608 609 612 616 617 618 619 621 640 651 652 655 659 665 666 675 683 684 685 686 692 693 694 695 716 717 730 742 743 792 819 851 852 879 880 890 892 908 938 |

Number of Samples | Max (%) | Min (%) | Mean (%) | Standard Deviation | |
---|---|---|---|---|---|

Training set | 122 | 87.027 | 4.477 | 58.454 | 20.523 |

Validation set | 41 | 86.475 | 6.087 | 52.714 | 22.762 |

Testing set | 41 | 87.937 | 7.415 | 51.007 | 24.220 |

Modeling Methods | Preprocessing Method | Dimensionality Reduction Method | R^{2}C | RMSEC | R^{2}P | RMSEP |
---|---|---|---|---|---|---|

RF | No | No | 0.927 | 5.884 | 0.888 | 7.395 |

PCA | 0.961 | 4.293 | 0.940 | 5.418 | ||

CARS | 0.934 | 5.566 | 0.905 | 6.815 | ||

MSC | No | 0.923 | 6.018 | 0.898 | 7.059 | |

PCA | 0.934 | 5.564 | 0.921 | 6.200 | ||

CARS | 0.915 | 6.333 | 0.920 | 6.238 | ||

SNV | No | 0.961 | 4.301 | 0.939 | 5.437 | |

PCA | 0.988 | 2.385 | 0.925 | 6.051 | ||

CARS | 0.935 | 5.554 | 0.902 | 6.914 | ||

SG | No | 0.986 | 2.524 | 0.930 | 5.834 | |

PCA | 0.986 | 2.505 | 0.962 | 4.311 | ||

CARS | 0.936 | 5.480 | 0.916 | 6.393 | ||

PSO-SVR | No | No | 0.903 | 7.017 | 0.889 | 6.832 |

PCA | 0.946 | 5.203 | 0.931 | 5.372 | ||

CARS | 0.951 | 4.982 | 0.943 | 4.876 | ||

MSC | No | 0.972 | 3.714 | 0.944 | 4.846 | |

PCA | 0.971 | 3.782 | 0.943 | 4.914 | ||

CARS | 0.965 | 4.217 | 0.944 | 4.868 | ||

SNV | No | 0.982 | 3.016 | 0.949 | 4.630 | |

PCA | 0.969 | 3.931 | 0.943 | 4.886 | ||

CARS | 0.949 | 5.068 | 0.943 | 4.890 | ||

SG | No | 0.902 | 7.039 | 0.889 | 6.848 | |

PCA | 0.984 | 2.775 | 0.971 | 3.448 | ||

CARS | 0.967 | 4.044 | 0.945 | 4.825 | ||

XGBoost | No | No | 0.991 | 1.933 | 0.917 | 6.867 |

PCA | 0.994 | 1.499 | 0.948 | 5.448 | ||

CARS | 0.990 | 2.001 | 0.911 | 7.129 | ||

MSC | No | 0.996 | 1.272 | 0.916 | 6.911 | |

PCA | 0.993 | 1.687 | 0.948 | 5.402 | ||

CARS | 0.991 | 2.000 | 0.928 | 6.396 | ||

SNV | No | 0.999 | 0.044 | 0.891 | 7.893 | |

PCA | 0.987 | 2.360 | 0.944 | 5.612 | ||

CARS | 0.991 | 1.932 | 0.923 | 6.615 | ||

SG | No | 0.991 | 1.897 | 0.911 | 7.100 | |

PCA | 0.994 | 1.548 | 0.946 | 5.549 | ||

CARS | 0.993 | 1.662 | 0.928 | 6.380 |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Yu, H.; Hu, Y.; Qi, L.; Zhang, K.; Jiang, J.; Li, H.; Zhang, X.; Zhang, Z.
Hyperspectral Detection of Moisture Content in Rice Straw Nutrient Bowl Trays Based on PSO-SVR. *Sustainability* **2023**, *15*, 8703.
https://doi.org/10.3390/su15118703

**AMA Style**

Yu H, Hu Y, Qi L, Zhang K, Jiang J, Li H, Zhang X, Zhang Z.
Hyperspectral Detection of Moisture Content in Rice Straw Nutrient Bowl Trays Based on PSO-SVR. *Sustainability*. 2023; 15(11):8703.
https://doi.org/10.3390/su15118703

**Chicago/Turabian Style**

Yu, Haiming, Yuhui Hu, Lianxing Qi, Kai Zhang, Jiwen Jiang, Haiyuan Li, Xinyue Zhang, and Zihan Zhang.
2023. "Hyperspectral Detection of Moisture Content in Rice Straw Nutrient Bowl Trays Based on PSO-SVR" *Sustainability* 15, no. 11: 8703.
https://doi.org/10.3390/su15118703