# Study on the Optimization of Hyperspectral Characteristic Bands Combined with Monitoring and Visualization of Pepper Leaf SPAD Value

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Sample Collection

#### 2.2. Chlorophyll Determination

^{2}; measurement accuracy: ±1.0 SPAD unit; and measurement range: −9.9–199.9 SPAD unit. Each leaf was divided into six plots (as shown in Figure 2). Three measurements were recorded for each plot, and the average value was taken as the final result of the chlorophyll content of the leaves.

#### 2.3. Hyperspectral Data Collection

#### 2.4. Spectral Extraction

#### 2.5. Research Methods

#### 2.5.1. Correlation Coefficient Method

#### 2.5.2. Stability Competitive Adaptive Reweighted Sampling (sCARS)

- Select N wavelength subsets from N Monte Carlo sampling [19] runs in an iterative and competitive manner. In each sampling process, a fixed proportion of samples is randomly selected to establish a calibration model.
- Perform a two-step process to select characteristic wavelengths: use an exponential decrease function [17] for wavelength selection and use adaptive reweighted sampling to achieve competitive wavelength selection.
- Use cross-validation [20] to select the subset with the smallest cross-validation root mean square error (RMSECV).

#### 2.5.3. Iteratively Retaining Informative Variables

#### 2.5.4. Partial Least-Squares Regression

#### 2.5.5. Extreme Gradient Boosting (XGBoost)

#### 2.5.6. Random Forest Regression (RFR)

#### 2.5.7. Gradient Boosting Decision Tree (GBDT)

#### 2.6. Accuracy Evaluation

#### 2.7. Technical Roadmap

## 3. Results

#### 3.1. Selection of Characteristic Band Based on CA Algorithm

#### 3.2. Selection of Characteristic Band Based on SCARS Algorithm

#### 3.3. Selection of Characteristic Band Based on IRIV Algorithm

#### 3.4. Screening Results

#### 3.5. Optimal Algorithm Selection

#### 3.5.1. Accuracy Comparison of Different Methods

#### 3.5.2. Model Construction Based on the Bands selected by the IRIV Algorithm

#### 3.6. Chlorophyll Distribution of Pepper Leaves

#### 3.7. Statistical Summary Based on the IRIV-XGBoost Algorithm

## 4. Discussion

## 5. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Dordas, C. Nitrogen Nutrition Index and Leaf Chlorophyll Concentration and its relationship with Nitrogen Use Efficiency in Barley (Hordeum vulgare L.). J. Plant Nutr.
**2017**, 40, 1190–1203. [Google Scholar] - Ata-Ul-Karim, S.T.; Cao, Q.; Zhu, Y.; Tang, L.; Rehmani, M.I.A.; Cao, W. Non-destructive Assessment of Plant Nitrogen Parameters Using Leaf Chlorophyll Measurements in Rice. Front. Plant Sci.
**2016**, 7, 1829. [Google Scholar] - Liu, B.; Shen, W.; Yue, Y.M.; Li, R.; Tong, Q.; Zhang, B. Combining spatial and spectral information to estimate chlorophyll contents of crop leaves with a field imaging spectroscopy system. Precis. Agric.
**2017**, 18, 491–506. [Google Scholar] - Manley, M. Near-infrared spectroscopy and hyperspectral imaging: Non-destructive analysis of biological materials. Chem. Soc. Rev.
**2014**, 43, 8200. [Google Scholar] - Sun, H.; Zheng, T.; Liu, N.; Cheng, M.; Li, M.; Zhang, Q. Vertical distribution of chlorophyll in potato plants based on hyperspectral imaging. Trans. Chin. Soc. Agric. Eng.
**2018**, 34, 149–156. [Google Scholar] - Liu, D.; Sun, D.W.; Zeng, X.A. Recent Advances in Wavelength Selection Techniques for Hyperspectral Image Processing in the Food Industry. Food Bioprocess Technol.
**2014**, 7, 307–323. [Google Scholar] - Ji-Yong, S.; Xiao-Bo, Z.; Jie-Wen, Z.; Kai-Liang, W. Nondestructive diagnostics of nitrogen deficiency by cucumber leaf chlorophyll distribution map based on near infrared hyperspectral imaging. Sci. Horticult.
**2012**, 138, 190–197. [Google Scholar] - Zhao, T.; Nakano, A.; Iwaski, Y.; Umeda, H. Application of Hyperspectral Imaging for Assessment of Tomato Leaf Water Status in Plant Factories. Appl. Sci.
**2020**, 10, 4665. [Google Scholar] - Daughtry, C.; Walthall, C.L.; Kim, M.S.; Colstoun, E.; Iii, M.M. Estimating Corn Leaf Chlorophyll Concentration from Leaf and Canopy Reflectance. Remote Sens. Environ.
**2000**, 74, 229–239. [Google Scholar] - Wu, C.; Zheng, N.; Tang, Q.; Huang, W. Estimating chlorophyll content from hyperspectral vegetation indices: Modeling and validation. Agric. For. Meteorol.
**2008**, 148, 1230–1241. [Google Scholar] - Yu, K.Q.; Zhao, Y.R.; Li, X.L.; Shao, Y.N.; Fei, L.; Yong, H. Hyperspectral Imaging for Mapping of Total Nitrogen Spatial Distribution in Pepper Plant. PLoS ONE
**2014**, 9, e116205. [Google Scholar] - Liu, Y.; Fan, M. Application Feasibility of SPAD-502 in Diagnosis of Potato Nitrogen Nutrient Status. Chin. Potato J.
**2012**, 26, 45–48. [Google Scholar] - Coste, S.; Baraloto, C.; Leroy, C.; Marcon, É.; Renaud, A.; Richardson, A.D.; Roggy, J.C.; Schimann, H.; Uddling, J.; Hérault, B. Assessing foliar chlorophyll contents with the SPAD-502 chlorophyll meter: A calibration test with thirteen tree species of tropical rainforest in French Guiana. Ann. For. Sci.
**2010**, 67, 607. [Google Scholar] - Nau, J.; Prokopová, J.; Ebíek, J.; Pundová, M. SPAD chlorophyll meter reading can be pronouncedly affected by chloroplast movement. Photosynth. Res.
**2010**, 105, 265–271. [Google Scholar] - Wang, D.; Aixia, S.U.; Liu, W. Trend Analysis of Vegetation Cover Changes Based on Spearman Rank Correlation Coefficient. J. Appl. Sci.
**2019**, 37, 519–528. [Google Scholar] - Amarkhil, Q.; Elwakil, E.; Hubbard, B. A Meta-Analysis of Critical Causes of Project Delay Using Spearman’s Rank and Relative Importance Index Integrated Approach. Can. J. Civil. Eng.
**2020**, 48, 1498–1507. [Google Scholar] - Li, H.; Liang, Y.; Xu, Q.; Cao, D. Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration. Anal. Chim. Acta
**2009**, 648, 77–84. [Google Scholar] - Liu, G.H.; Xia, R.S.; Jiang, H.; Mei, C.-L.; Huanng, Y.-H. A Wavelength Selection Approach of Near Infrared Spectra Based on SCARS Strategy and Its Application. Spectrosc. Spectr. Anal.
**2014**, 34, 2094–2097. [Google Scholar] - Holmes, E.W.; Kahn, S.E.; Molnar, P.A.; Bermes, E.W. Verification of reference ranges by using a Monte Carlo sampling technique. Clin. Chem.
**2020**, 40, 2216–2222. [Google Scholar] - Guo, L.N.; Fan, G.S. Support Vector Machines for Surface Soil Density Prediction based on Grid Search and Cross Validation. Chin. J. Soil Sci.
**2018**, 49, 512–518. [Google Scholar] - Zhang, H.; Wang, Y.; Liu, C.; Jiang, H. Influence of surfactant CTAB on the electrochemical performance of manganese dioxide used as supercapacitor electrode material. J. Alloys Compd.
**2012**, 517, 8. [Google Scholar] - Yun, Y.H.; Wang, W.; Tan, M.; Liang, Y.; Li, H.; Cao, D.; Lu, H.; Xu, Q. A strategy that iteratively retains informative variables for selecting optimal variable subset in multivariate calibration. Anal. Chim. Acta
**2014**, 807, 36–43. [Google Scholar] - Zhang, H.; Wang, H.; Dai, Z.; Chen, M.S.; Yuan, Z. Improving accuracy for cancer classification with a new algorithm for genes selection. BMC Bioinform.
**2012**, 13, 298. [Google Scholar] - Mann, H.B.; Whitney, D.R. On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other. Ann. Math. Stat.
**1947**, 18, 50–60. [Google Scholar] - Nawar, S.; Buddenbaum, H.; Hill, J.; Kozak, J. Modeling and Mapping of Soil Salinity with Reflectance Spectroscopy and Landsat Data Using Two Quantitative Methods (PLSR and MARS). Remote Sens.
**2014**, 6, 10813–10834. [Google Scholar] [CrossRef] [Green Version] - Wang, S.; Chen, Y.; Wang, M.; Zhao, Y.; Li, J. SPA-Based Methods for the Quantitative Estimation of the Soil Salt Content in Saline-Alkali Land from Field Spectroscopy Data: A Case Study from the Yellow River Irrigation Regions. Remote Sens.
**2019**, 11, 967. [Google Scholar] - Weng, Y.; Qi, H.; Fang, H. PLSR-Based Hyperspectral Remote Sensing Retrieval of Soil Salinity of Chaka-Gonghe Basin in Qinghai Province. Acta Pedol. Sin.
**2010**, 47, 1255–1263. [Google Scholar] - Chen, T.; Guestrin, C. XGBoost: A Scalable Tree Boosting System. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 13–17 August 2016; pp. 785–794. [Google Scholar]
- Jia, Y.; Jin, S.; Savi, P.; Gao, Y.; Tang, J.; Chen, Y.; Li, W. GNSS-R Soil Moisture Retrieval Based on a XGboost Machine Learning Aided Method: Performance and Validation. Remote Sens.
**2019**, 11, 1655. [Google Scholar] - Loggenberg, K.; Strever, A.; Greyling, B.; Poona, N. Modelling Water Stress in a Shiraz Vineyard Using Hyperspectral Imaging and Machine Learning. Remote Sens.
**2018**, 10, 202. [Google Scholar] - Tong, Y.; Lu, K.; Yang, Y.; Li, J.; Qian, J. Can Natural Language Processing Help Differentiate Inflammatory Intestinal Diseases in China? Models Applying Random Forest and Convolutional Neural Network Approaches. BMC
**2020**, 20, 248. [Google Scholar] - Liu, L.; Ji, M.; Dong, Y.; Zhang, R.; Buchroithner, M. Quantitative Retrieval of Organic Soil Properties from Visible Near-Infrared Shortwave Infrared (Vis-NIR-SWIR) Spectroscopy Using Fractal-Based Feature Extraction. Remote Sens.
**2016**, 8, 1035. [Google Scholar] - Jin, X.; Zhu, X.; Li, S.; Wang, W.; Qi, H. Predicting Soil Available Phosphorus by Hyperspectral Regression Method Based on Gradient Boosting Decision Tree. Laser Optoelectron. Prog.
**2019**, 56. [Google Scholar] - Zhao, Z.H.; Zhang, C.Z.; Zhang, J.B.; Wu, Y.J.; Zhang, H.M.; Lu, C.Y. The Hyperspectral Inversion for Estimating Maize Chlorophyll Contents Based on Regression Analysis. Chin. Agric. Sci. Bull.
**2021**, 35, 7–16. [Google Scholar] - Sarathjith, M.C.; Das, B.S.; Wani, S.P.; Sahrawat, K.L. Variable indicators for optimum wavelength selection in diffuse reflectance spectroscopy of soils. Geoderma
**2016**, 267, 1–9. [Google Scholar] - Xu, S.; Zhao, Y.; Wang, M.; Shi, X. Determination of rice root density from Vis–NIR spectroscopy by support vector machine regression and spectral variable selection techniques. Catena
**2017**, 157, 12–23. [Google Scholar] - Pinto, C.F.; Parab, J.S.; Sequeira, M.D.; Naik, G.M. Development of Altera NIOS II Soft-core system to predict total Hemoglobin using Multivariate Analysis. J. Phys. Conf. Ser.
**2021**, 1921, 12039. [Google Scholar] - Wei, L.; Yuan, Z.; Yu, M.; Huang, C.; Cao, L. Estimation of Arsenic Content in Soil Based on Laboratory and Field Reflectance Spectroscopy. Sensors
**2019**, 19, 3904. [Google Scholar] - Raissi, M.; Karniadakis, G.E. Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations. J. Comput. Phys.
**2018**, 357, 125–141. [Google Scholar]

**Figure 3.**Schematic diagram of the GaiaSorter hyperspectral imaging system. 1. Hyperspectral imager, 2. imaging lens, 3. halogen lamp, 4. sample table, 5. correction whiteboard, and 6. electric translation table.

**Figure 7.**Characteristic variable selection process of sCARS algorithm. (

**a**) Changes in the number of waveband variables. (

**b**) Validation of RMSECV. (

**c**) Path of variable regression coefficients.

**Figure 8.**IRIV algorithm selection process: (

**a**) The change in the number of retained informative variables in each round; (

**b**) Changes in P value and DMEAN in the sixth round.

**Figure 10.**Scatter plot of measured and predicted values of the four models: (

**a**) PLSR; (

**b**) XGBoost; (

**c**) RFR; and (

**d**) GBDT.

**Figure 11.**Distribution of SPAD value in the lower leaf in different models: (

**a**,

**e**—PLSR), (

**b**,

**f**—XGBoost), (

**c**,

**g**—RFR), (

**d**,

**h**—GBDT).

**Figure 12.**Distribution of SPAD value in the middle leaf in different models: (

**a**,

**e**—PLSR), (

**b**,

**f**—XGBoost), (

**c**,

**g**—RFR), (

**d**,

**h**—GBDT).

**Figure 13.**Distribution of SPAD value in the upper leaf in different models: (

**a**,

**e**—PLSR), (

**b**,

**f**—XGBoost), (

**c**,

**g**—RFR), (

**d**,

**h**—GBDT).

Wavelength Variable Type | Classification Rules |
---|---|

Strongly informative | $DMEA{N}_{i}<0$$,{P}_{i}0.05$ |

Weakly informative | $DMEA{N}_{i}<0$$,{P}_{i}0.05$ |

Uninformative | $DMEA{N}_{i}>0$$,{P}_{i}0.05$ |

Interfering | $DMEA{N}_{i}>0$$,{P}_{i}0.05$ |

Selection Method | Number of Bands | Modeling Algorithm | ${\mathit{R}}_{\mathit{c}\mathit{v}}^{2}$ | $\mathit{R}\mathit{M}\mathit{S}{\mathit{E}}_{\mathit{c}\mathit{v}}$ | $\mathit{M}\mathit{A}{\mathit{E}}_{\mathit{c}\mathit{v}}$ |
---|---|---|---|---|---|

Full bands | 176 | PLSR | 0.52 | 2.57 | 2.11 |

XGBoost | 0.48 | 2.80 | 2.28 | ||

RFR | 0.42 | 2.95 | 2.83 | ||

GBDT | 0.50 | 2.76 | 2.19 | ||

CA | 76 | PLSR | 0.48 | 2.59 | 2.1 |

XGBoost | 0.29 | 3.00 | 2.39 | ||

RFR | 0.41 | 2.95 | 2.4 | ||

GBDT | 0.44 | 2.84 | 2.23 | ||

sCARS | 46 | PLSR | 0.55 | 2.59 | 2.13 |

XGBoost | 0.54 | 2.68 | 2.17 | ||

RFR | 0.43 | 2.92 | 2.32 | ||

GBDT | 0.53 | 2.74 | 2.17 | ||

IRIV | 26 | PLSR | 0.84 | 2.46 | 2.02 |

XGBoost | 0.81 | 2.76 | 2.30 | ||

RFR | 0.80 | 2.85 | 2.28 | ||

GBDT | 0.80 | 2.82 | 2.22 |

**Table 3.**Statistical information of chlorophyll inversion map of pepper leaves under different models and different leaf positions.

Leaf Position | Measured Value | Model Method | Min Value | Max Value |
---|---|---|---|---|

Lower leaf | 66.0 | PLSR | 19 | 82 |

XGBoost | 43 | 69 | ||

RFR | 46 | 67 | ||

GBDT | 43 | 70 | ||

69.0 | PLSR | 21 | 85 | |

XGBoost | 44 | 69 | ||

RFR | 47 | 67 | ||

GBDT | 45 | 69 | ||

Middle leaf | 61.0 | PLSR | 14 | 82 |

XGBoost | 42 | 69 | ||

RFR | 45 | 66 | ||

GBDT | 42 | 69 | ||

60.6 | PLSR | 12 | 83 | |

XGBoost | 41 | 69 | ||

RFR | 45 | 66 | ||

GBDT | 44 | 69 | ||

Upper leaf | 48.3 | PLSR | 3 | 73 |

XGBoost | 42 | 67 | ||

RFR | 45 | 65 | ||

GBDT | 42 | 68 | ||

50.5 | PLSR | 2 | 72 | |

XGBoost | 42 | 67 | ||

RFR | 45 | 64 | ||

GBDT | 42 | 68 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2021 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**

Yuan, Z.; Ye, Y.; Wei, L.; Yang, X.; Huang, C.
Study on the Optimization of Hyperspectral Characteristic Bands Combined with Monitoring and Visualization of Pepper Leaf SPAD Value. *Sensors* **2022**, *22*, 183.
https://doi.org/10.3390/s22010183

**AMA Style**

Yuan Z, Ye Y, Wei L, Yang X, Huang C.
Study on the Optimization of Hyperspectral Characteristic Bands Combined with Monitoring and Visualization of Pepper Leaf SPAD Value. *Sensors*. 2022; 22(1):183.
https://doi.org/10.3390/s22010183

**Chicago/Turabian Style**

Yuan, Ziran, Yin Ye, Lifei Wei, Xin Yang, and Can Huang.
2022. "Study on the Optimization of Hyperspectral Characteristic Bands Combined with Monitoring and Visualization of Pepper Leaf SPAD Value" *Sensors* 22, no. 1: 183.
https://doi.org/10.3390/s22010183