# Multispectral UAV-Based Monitoring of Leek Dry-Biomass and Nitrogen Uptake across Multiple Sites and Growing Seasons

^{1}

^{2}

^{3}

^{4}

^{5}

^{6}

^{*}

## Abstract

**:**

_{ct}= 6.60 g plant

^{−1}, R

^{2}= 0.90). Leek N-uptake was predicted most accurately by a simple linear regression model based on the red wide dynamic range (RWDRVI) (RMSE

_{ct}= 0.22 gN plant

^{−1}, R

^{2}= 0.85). The results showed that randomized Kfold-CV is an undesirable approach. It resulted in more consistent and lower RMSE values during model training and selection, but worse performance on new data. This would be due to information leakage of flight-specific conditions in the validation data split. However, the model predictions were less accurate for data acquired in a different growing season (DBM: RMSEP = 8.50 g plant

^{−1}, R

^{2}= 0.77; N-uptake: RMSEP = 0.27 gN plant

^{−1}, R

^{2}= 0.68). Recalibration might solve this issue, but additional research is required to cope with this effect during image acquisition and processing. Further improvement of the model robustness could be obtained through the inclusion of phenological parameters such as crop height.

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Data Acquisition

#### 2.1.1. Site Description and Agricultural Management

#### 2.1.2. Plant Samples

^{−1}). The total nitrogen content was determined from the plant samples using the Dumas combustion method, expressed as gN kg

^{−1}fresh weight [26]. The leek nitrogen uptake was then determined by multiplying this number by the average fresh weight of the harvest.

#### 2.1.3. UAV Monitoring

#### 2.2. Image Processing and Feature Extraction

#### 2.3. Statistical Analysis

#### 2.3.1. Modelling Approaches

^{−1}) and nitrogen uptake (gN plant

^{−1}). In total, six different modelling approaches were studied because they possess various mathematical properties. The linear models consisted of simple linear regression (SLR), lasso regression (lasso) and partial least squares regression (PLSR). Support vector regression (SVR), random forest regression (RFR) and gradient boosting with a tree-based model as a base estimator (xGBR) were studied as non-linear models. The SLR models contained only one predictor variable. The other models were more complex as they included the median and iqr of all VIs. A lasso regression model automatically selects predictor variables [39] while a PLSR model reduces the dimensionality by projecting the predictors onto a lower-dimensional space where the latent variables are orthogonal [40]. An SVR model determines a hyperplane (i.e., support vectors) that captures the relationship between the predictor variables and target variable [41]. When the Euclidean distance of a data point to a support vector is below a threshold (ε), the data point is not considered in calculating the model performance index. Finally, random forest regression (RFR) and extreme gradient boosting regression (xGBR) are both ensemble regression methods that predict the target values based on a stack of multiple regression trees [42,43]. RFR uses numerous trees fitted on bootstrapped datasets and finally aggregates the predictions of all trees to predict the response value of a new observation. In an xGBR model, the first decision tree is fitted on the complete dataset, and the following trees are built on a dataset using the residual of the previous tree as the response. The SLR, Lasso, PLSR, SVR and RFR models were fitted using python’s Scikit-Learn (v.0.22.1) package [44], while the xGBR models were estimated with the XGBoost (v.1.2.0) package [42].

#### 2.3.2. Model Training and Hyperparameter Optimization

_{cv}) (Equation (2)) on the training data. So, the data of the remaining n−1 flights to train the model were iteratively split in a training set of n−2 flights and a validation set containing all data from the remaining flight. For the PLSR model, the number of latent variables was selected as the number with the lowest RMSE

_{cv}. For the lasso model, the regularization parameter was optimized. Before the estimation of the Lasso or PLSR model, the predictor variables were scaled to zero mean and unit standard deviation. For the SVR model, only the RBF kernel was considered. The kernel scaling factor (gamma), ε and L2-regularization parameters were optimized. For the RFR model, the number of regression trees, the minimal number of observations at each leaf, the maximal number of predictors used and the maximal depth of each tree were optimized. For the xGBR model, the following hyperparameters were tuned: the number of regression trees that were used, the maximal tree depth, the minimum reduction in the loss function value, learning rate, L2 regularization weight and the number of predictors used to build the trees. The hyperparameters for the SVR, RFR and xGBR models were optimized through Bayesian optimization using the scikit-optimize package (v0.8.1).

_{ct}, Equation (2)).

#### 2.3.3. Model Selection and Validation

^{2}of the linear regression model between the observed and predicted values. The $RMS{E}_{ct}$ was calculated as the average error across all LOF-CV iterations as the number of observations per flight varied from 8 to 24.

_{i}the number of observations in flight i, and ${\widehat{y}}_{ij}$ the jth observation and prediction of flight i, respectively, and σ

_{obs}the standard deviation of the observations.

## 3. Results

#### 3.1. UAV Flights and Plant Samples

^{−1}and 29.25 g plant

^{−1}for the 2019 and 2020 experiments, respectively. The minimal and maximal measured DBM values were 0.71 and 70.15 in 2019, and 1.34 and 97.98 g plant

^{−1}in 2020. The observed leek nitrogen uptake in 2019 ranged from 0.02 to 2.16 with a mean value of 0.94 gN plant

^{−1}and in 2020, nitrogen uptake ranged from 0.04 to 1.62, with a mean value of 0.78 gN plant

^{−1}.

#### 3.2. Regression Analysis and Model Selection

#### 3.2.1. SLR Models

_{ct}of 9.08 g plant

^{−1}(Figure 5a). The median NDVI and BNDVI had only a slightly increased RMSE

_{ct}, 9.37 and 9.96 g plant

^{−1}, respectively. The N-uptake was also most accurately predicted by the RWDRVI model with an RMSE

_{ct}of 0.22 gN plant

^{−1}. The SLR models using the median NDVI, BNDVI, GNDVI and BWDRVI had a similar accuracy (Figure 5b). The RMSE

_{ct}only varied from 0.23 to 0.25 gN plant

^{−1}. The N-uptake and DBM were both least accurately estimated using the iqr of the NDRE; the RMSE

_{ct}was equal to 0.67 gN plant

^{−1}and 25.26 g/plant, respectively. However, the relationship between the observed and predicted values in LOF-CV was significant (p < 0.001). Only the iqr of the MTCI index resulted in a non-significant linear model for both DBM (p = 0.05) and N-uptake (p = 0.08) estimation. More complex models were trained and compared to the best performing SLR model to evaluate whether the information contributed by multiple VIs could provide more accurate predictions.

#### 3.2.2. Model Selection for DBM Prediction

^{−1}. The other models had smaller training errors that varied from 1.97 g plant

^{−1}for the xGBR model to 5.26 g plant

^{−1}for the PLSR model. However, the average RMSE

_{cv}of the internal cross-validation loop (Figure 2) for the SLR model was only 6.34 g plant

^{−1}, which was similar to the ${\overline{RMSE}}_{cv}$ of the RFR and xGBR models, which had the lowest values of the multivariate models.

_{ct}value (6.60 g plant

^{−1}) and large R

^{2}(0.9). Additionally, its RMSE values during training, cross-validation and cross-testing were relatively consistent. Although the RMSE

_{ct}of the Lasso model was slightly larger than for the RFR model, the Lasso model was preferred, as the model is considerably less complex than the RFR model. Besides, the RMSE

_{ct}value of the RFR model was almost two times larger than the ${\overline{RMSE}}_{train}$, which indicates the model was prone to overfitting. The results showed that the SVR and xGBR models had the same issue: a small ${\overline{RMSE}}_{train}$ but considerably larger RMSE

_{ct}values.

#### 3.2.3. Model Selection for N-Uptake Prediction

^{−1}. The SLR model performed the least accurately during model training, with an average training error of 0.21 gN plant

^{−1}. However, the SLR model was the second most accurate model in cross-validation and cross-testing with an RMSE

_{cv}and RMSE

_{ct}of 0.20 and 0.22 gN plant

^{−1}, respectively. Additionally, the SLR model had consistent RMSE values during training, internal cross-validation, and cross-testing. The bias of the SLR model was also almost zero. Although the SVR model was the best performing model during training, the RMSEcv (0.30 gN plant

^{−1}) was much larger than the ${\overline{RMSE}}_{train}$ (0.08 gN plant

^{−1}). The SVR model had also the largest bias compared to the other models.

_{pearson}= 0.44) and N-uptake (r

_{pearson}= 0.66) values was less clear.

#### 3.3. Model Validation

#### 3.3.1. Pluston Variety

^{−1}) was observed (+28.8%) compared to the RMSE

_{ct}(Table 5). Additionally, the R

^{2}decreased (0.77) and absolute bias (1.59 g plant

^{−1}) increased compared to cross-testing. DBM was less accurately predicted at the farmers’ fields compared to the research centres as the bias and RMSEP increased to 3.11 g plant

^{−1}and 9.79 g plant

^{−1}, respectively. The prediction error was relatively small for low DBM values but increased for larger DBM values. As the maximal observed DBM in 2020 was smaller than in 2019, no clear underestimation of the DBM yield was observed (Figure 7a). Similar trends were observed for the validation of the SLR model to predict the Pluston N-uptake. The RMSEP increased 22.7% relative to the RMSE

_{ct}to a value of 0.27 gN plant

^{−1}, while also the R

^{2}(0.68) decreased and the absolute bias increased slightly to 0.03 gN plant

^{−1}. N-uptake in the beginning of the growing season was clearly underestimated at the research centres, while this was not the case for the farmer fields. The model predictions were less accurate for large N-uptake values (Figure 7b).

#### 3.3.2. Model Transfer to Other Varieties

^{−1}(Table 6). The Harston DBM was the least accurately predicted with an RMSEP of 21.68 g plant

^{−1}. Low DBM values were largely overestimated while a very large DBM yield was observed, 97.98 g plant

^{−1}, which was clearly underestimated. The RWDRVI was able to predict the N-uptake of the Harston and Krypton varieties with similar accuracy compared to Pluston, the RMSEP was 0.26 gN plant

^{−1}for both varieties, while the bias was, respectively −0.01 and 0.04 gN plant

^{−1}(Table 6). The RMSEP of the Chiefton variety was 0.27 gN plant

^{−1}, which could be largely attributed to the large bias of 0.24 gN plant

^{−1}. The N-uptake of the Vitaton variety was least accurately predicted with an RMSEP of 0.44 gN plant

^{−1}and a bias of 0.29 gN plant

^{−1}.

## 4. Discussion

#### 4.1. Model Selection and Validation

_{ct}(Table A1 and Table A2) is compared to RMSEP (Table 5).

#### 4.2. Model Application

## 5. Conclusions

_{ct}were observed. A Lasso model, using RWDRVI and MCARI resulting in an RMSE

_{ct}of 6.60 g plant

^{−1}, a bias of 0.57 g plant

^{−1}and an R

^{2}of 0.90 was determined as the best model to predict leek DBM. An SLR using the RWDRVI resulted in an RMSE

_{ct}of 0.22 gN plant

^{−1}, a bias

_{ct}of 0.00 gN plant and an R

^{2}of 0.85. The models still have some limitations as the model predictions were less accurate when used to predict DBM and N-uptake during the next growing season, as the RMSEP and bias increased while the R

^{2}decreased. This study also showed that Kfold-CV could lead to over-optimistic results and the consequent selection of models with poor generalizability due to the inclusion of flight-specific effects such as the weather conditions, soil type, and sensor calibration in the validation data. Additional research into these effects is recommend to build models that are capable of predicting DBM and N-uptake accurately throughout different growing seasons and fields from multispectral UAV data. As long as this has not been resolved, field- and season-specific calibration will be required to improve model accuracy, which impedes the applicability of the technology.

## Author Contributions

## Funding

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## Appendix A

**Table A1.**Model training performance using randomized Kfold-CV for the prediction of the leek DBM using the data from the 2019 experiments.

SLR | Lasso | PLSR | SVR | RFR | xGBR | |
---|---|---|---|---|---|---|

${\overline{RMSE}}_{train}$ (g plant^{−1}) | 8.15 | 4.49 | 5.07 | 3.58 | 1.7 | 1.96 |

${\overline{RMSE}}_{cv}$ (g plant^{−1}) | 8.17 | 4.88 | 5.32 | 4.36 | 4.18 | 4.35 |

RMSE_{ct} (g plant^{−1}) | 8.19 | 4.47 | 5.38 | 4.31 | 4.20 | 4.52 |

rRMSE_{ct} (%) | 38.74 | 23.05 | 25.47 | 20.0 | 19.89 | 21.38 |

Bias_{ct} (g plant^{−1}) | 0.01 | 0.00 | 0.10 | −0.05 | −0.20 | −0.63 |

R^{2} | 0.85 | 0.95 | 0.94 | 0.96 | 0.96 | 0.96 |

**Table A2.**Model training performance using randomized Kfold-CV for the prediction of the leek N-uptake using the data from the 2019 experiments.

SLR | Lasso | PLSR | SVR | RFR | xGBR | |
---|---|---|---|---|---|---|

${\overline{RMSE}}_{train}$ (g plant^{−1}) | 0.21 | 0.15 | 0.14 | 0.08 | 0.06 | 0.14 |

${\overline{RMSE}}_{cv}$ (g plant^{−1}) | 0.21 | 0.16 | 0.16 | 0.16 | 0.14 | 0.15 |

RMSE_{ct} (g plant^{−1}) | 0.21 | 0.16 | 0.16 | 0.17 | 0.14 | 0.16 |

rRMSE_{ct} (%) | 36.44 | 28.36 | 27.19 | 29.23 | 24.35 | 27.83 |

Bias_{ct} (g plant^{−1}) | 0.00 | 0.00 | 0.00 | 0.01 | 0.00 | 0.01 |

R^{2} | 0.87 | 0.92 | 0.93 | 0.91 | 0.94 | 0.93 |

## References

- Zhang, X.; Davidson, E.A.; Mauzerall, D.L.; Searchinger, T.D.; Dumas, P.; Shen, Y. Managing Nitrogen for Sustainable Development. Nature
**2015**, 528, 51–59. [Google Scholar] [CrossRef][Green Version] - Thompson, R.B.; Tremblay, N.; Fink, M.; Gallardo, M.; Padilla, F.M. Tools and Strategies for Sustainable Nitrogen Fertilisation of Vegetable Crops; Springer: Cham, Switzerland, 2017; pp. 11–63. [Google Scholar]
- Groenten Openlucht|Landbouw & Visserij. Available online: https://landbouwcijfers.vlaanderen.be/landbouw/groenten-openlucht (accessed on 31 August 2022).
- FAO Crops and Livestock Products. Available online: https://www.fao.org/faostat/en/#data/QCL (accessed on 31 August 2022).
- Thompson, R.B.; Voogt, W.; Incrocci, L.; Fink, M.; de Neve, S. Strategies for Optimal Fertiliser Management of Vegetable Crops in Europe. Acta Hortic.
**2018**, 1192, 129–140. [Google Scholar] [CrossRef] - Tei, F.; de Neve, S.; de Haan, J.; Kristensen, H.L. Nitrogen Management of Vegetable Crops. Agric. Water Manag.
**2020**, 240, 106316. [Google Scholar] - Padilla, F.M.; Gallardo, M.; Peña-Fleitas, M.T.; de Souza, R.; Thompson, R.B. Proximal Optical Sensors for Nitrogen Management of Vegetable Crops: A Review. Sensors
**2018**, 18, 2083. [Google Scholar] [CrossRef] [PubMed][Green Version] - Ulissi, V.; Antonucci, F.; Benincasa, P.; Farneselli, M.; Tosti, G.; Guiducci, M.; Tei, F.; Costa, C.; Pallottino, F.; Pari, L.; et al. Nitrogen Concentration Estimation in Tomato Leaves by VIS-NIR Non-Destructive Spectroscopy. Sensors
**2011**, 11, 6411–6424. [Google Scholar] [CrossRef][Green Version] - de Souza, R.; Peña-Fleitas, M.T.; Thompson, R.B.; Gallardo, M.; Grasso, R.; Padilla, F.M. The Use of Chlorophyll Meters to Assess Crop N Status and Derivation of Sufficiency Values for Sweet Pepper. Sensors
**2019**, 19, 2949. [Google Scholar] [CrossRef] [PubMed][Green Version] - de Souza, R.; Grasso, R.; Teresa Peña-Fleitas, M.; Gallardo, M.; Thompson, R.B.; Padilla, F.M. Effect of Cultivar on Chlorophyll Meter and Canopy Reflectance Measurements in Cucumber. Sensors
**2020**, 20, 509. [Google Scholar] [CrossRef][Green Version] - Corti, M.; Marino Gallina, P.; Cavalli, D.; Cabassi, G. Hyperspectral Imaging of Spinach Canopy under Combined Water and Nitrogen Stress to Estimate Biomass, Water, and Nitrogen Content. Biosyst. Eng.
**2017**, 158, 38–50. [Google Scholar] [CrossRef] - Marino, S.; Alvino, A. Hyperspectral Vegetation Indices for Predicting Onion (Allium cepa L.) Yield Spatial Variability. Comput. Electron. Agric.
**2015**, 116, 109–117. [Google Scholar] [CrossRef] - Suarez, L.A.; Robson, A.; McPhee, J.; O’Halloran, J.; van Sprang, C. Accuracy of Carrot Yield Forecasting Using Proximal Hyperspectral and Satellite Multispectral Data. Precis. Agric.
**2020**, 21, 1304–1326. [Google Scholar] [CrossRef] - Marino, S.; Basso, B.; Leone, A.P.; Alvino, A. Agronomic Traits and Vegetation Indices of Two Onion Hybrids. Sci. Hortic.
**2013**, 155, 56–64. [Google Scholar] [CrossRef] - Tsouros, D.C.; Bibi, S.; Sarigiannidis, P.G. A Review on UAV-Based Applications for Precision Agriculture. Information
**2019**, 10, 349. [Google Scholar] [CrossRef][Green Version] - Wittstruck, L.; Kühling, I.; Trautz, D.; Kohlbrecher, M.; Jarmer, T. UAV-Based RGB Imagery for Hokkaido Pumpkin (Cucurbita Max.) Detection and Yield Estimation. Sensors
**2021**, 21, 118. [Google Scholar] [CrossRef] [PubMed] - Ballesteros, R.; Ortega, J.F.; Hernandez, D.; Moreno, M.A. Onion Biomass Monitoring Using UAV-Based RGB Imaging. Precis. Agric.
**2018**, 19, 840–857. [Google Scholar] [CrossRef] - Moeckel, T.; Dayananda, S.; Nidamanuri, R.R.; Nautiyal, S.; Hanumaiah, N.; Buerkert, A.; Wachendorf, M. Estimation of Vegetable Crop Parameter by Multi-Temporal UAV-Borne Images. Remote Sens.
**2018**, 10, 805. [Google Scholar] [CrossRef][Green Version] - Kim, D.-W.; Yun, H.; Jeong, S.-J.; Kwon, Y.-S.; Kim, S.-G.; Lee, W.; Kim, H.-J. Modeling and Testing of Growth Status for Chinese Cabbage and White Radish with UAV-Based RGB Imagery. Remote Sens.
**2018**, 10, 563. [Google Scholar] [CrossRef][Green Version] - Astor, T.; Dayananda, S.; Nautiyal, S.; Wachendorf, M. Vegetable Crop Biomass Estimation Using Hyperspectral and RGB 3D UAV Data. Agronomy
**2020**, 10, 1600. [Google Scholar] [CrossRef] - Xie, C.; Yang, C. A Review on Plant High-Throughput Phenotyping Traits Using UAV-Based Sensors. Comput. Electron. Agric.
**2020**, 178, 105731. [Google Scholar] [CrossRef] - Panday, U.S.; Pratihast, A.K.; Aryal, J.; Kayastha, R.B. A Review on Drone-Based Data Solutions for Cereal Crops. Drones
**2020**, 4, 41. [Google Scholar] [CrossRef] - Johansen, K.; Morton, M.J.L.; Malbeteau, Y.; Aragon, B.; Al-Mashharawi, S.; Ziliani, M.G.; Angel, Y.; Fiene, G.; Negrão, S.; Mousa, M.A.A.; et al. Predicting Biomass and Yield in a Tomato Phenotyping Experiment Using UAV Imagery and Random Forest. Front. Artif. Intell.
**2020**, 3, 28. [Google Scholar] [CrossRef] - de Nies, J.; Verhaeghe, M. Het Documenteren en Milieukundig Bijstellen van Het KNS en Andere Bemestingsadviessystemen in de Tuinbouw met het Oog op Een Ruimere Toepassing in de Tuinbouw Zoals Voorzien in het Actieprogramma 2011–2014. Available online: http://www.vlm.be/nl//SiteCollectionDocuments/Mestbank/Studies/Bemestingsadviessystementuinbouw/20141114eindrapportVlaamsKNS.pdf (accessed on 23 January 2020).
- Lorenz, H.P.; Schlaghecken, J.; Engel, G.; Maync, A.; Ziegler, J.; Strohmeyer, K. Ordnungsgemaesse Stickstoff-Versorgung im Freiland-Gemuesebau nach dem, Kulturbegleitenden-Nmin-Sollwerte-(KNS)-System; Ministerium fuer Landwirtschaft, Weinbau und Forsten Rheinland-Pfalz: Mainz, Germany, 1989; 85p. (In German) [Google Scholar]
- Mihaljev, Ž.A.; Jakšić, S.M.; Prica, N.B.; Ćupić, Ž.N.; Živkov-Baloš, M.M. Comparison of the Kjeldahl Method, Dumas Method and NIR Method for Total Nitrogen Determination in Meat and Meat Products. J. Agroaliment. Process. Technol.
**2015**, 21, 365–370. [Google Scholar] - Gillies, S.; Ward, B.; Petersen, A.S. Others Rasterio: Geospatial Raster I/O for Python Programmers. 2013. Available online: https://github.com/rasterio/rasterio (accessed on 7 October 2022).
- Haumont, J.; Lootens, P.; Cool, S.; van Beek, J.; Raymaekers, D.; Ampe, E.M.; de Cuypere, T.; Bes, O.; Bodyn, J.; Saeys, W. 60. Leek Growth Monitoring Using Multispectral UAV Imagery; Wageningen Academic Publishers: Wageningen, The Netherlands, 2021; pp. 501–507. [Google Scholar] [CrossRef]
- Rouse, J.W., Jr.; Haas, R.H.; Schell, J.A.; Deering, D.W. Monitoring Vegetation Systems in the Great Plains with Erts; NASA Special Publication: Washington, DC, USA, 1974; Volume 351, p. 309. [Google Scholar]
- Gitelson, A.; Merzlyak, M.N. Quantitative Estimation of Chlorophyll-a Using Reflectance Spectra: Experiments with Autumn Chestnut and Maple Leaves. J. Photochem. Photobiol. B
**1994**, 22, 247–252. [Google Scholar] [CrossRef] - Gitelson, A.; Kaufman, Y.; Merzlyak, M. Use of a Green Channel in Remote Sensing of Global Vegetation from EOS-MODIS. Remote Sens. Environ.
**1996**, 58, 289–298. [Google Scholar] [CrossRef] - Yang, C.; Everitt, J.H.; Bradford, J.M.; Murden, D. Airborne Hyperspectral Imagery and Yield Monitor Data for Mapping Cotton Yield Variability. Precis. Agric.
**2004**, 5, 445–461. [Google Scholar] [CrossRef] - Gitelson, A.A. Wide Dynamic Range Vegetation Index for Remote Quantification of Biophysical Characteristics of Vegetation. J. Plant Physiol.
**2004**, 161, 165–173. [Google Scholar] [CrossRef] [PubMed][Green Version] - Dash, J.; Curran, P.J. The MERIS Terrestrial Chlorophyll Index. Int. J. Remote Sens.
**2010**, 25, 5403–5413. [Google Scholar] [CrossRef] - Daughtry, C.S.T.; Walthall, C.L.; Kim, M.S.; de Colstoun, E.B.; McMurtrey, J.E. Estimating Corn Leaf Chlorophyll Concentration from Leaf and Canopy Reflectance. Remote Sens. Environ.
**2000**, 74, 229–239. [Google Scholar] [CrossRef] - Barnes, E.M.; Clarke, T.R.; Richards, S.E.; Colaizzi, P.D.; Haberland, J.; Kostrzewski, M.; Waller, P.; Choi, C.; Riley, E.; Thompson, T.; et al. Coincident Detection of Crop Water Stress, Nitrogen Status and Canopy Density Using Ground Based Multispectral Data. In Proceedings of the Fifth International Conference on Precision Agriculture, Bloomington, MN, USA, 16–19 July 2000; Volume 1619, p. 6. [Google Scholar]
- Gitelson, A.; Gritz, Y.; Merzlyak, M. Relationships between Leaf Chlorophyll Content and Spectral Reflectance and Algorithms for Non-Destructive Chlorophyll Assessment in Higher Plant Leaves. J. Plant Physiol.
**2003**, 160, 271–282. [Google Scholar] [CrossRef] - Huete, A.R. A Soil-Adjusted Vegetation Index (SAVI). Remote Sens. Environ.
**1988**, 25, 295–309. [Google Scholar] [CrossRef] - Tibshirani, R. Regression Shrinkage and Selection via the Lasso. J. R. Stat. Soc. Ser. B (Methodol.)
**1996**, 58, 267–288. [Google Scholar] [CrossRef] - Geladi, P.; Kowalski, B.R. Partial Least-Squares Regression: A Tutorial. Anal. Chim. Acta
**1986**, 185, 1–17. [Google Scholar] [CrossRef] - Smola, A.J.; Schölkopf, B. A Tutorial on Support Vector Regression. Stat. Comput.
**2004**, 14, 199–222. [Google Scholar] [CrossRef][Green Version] - 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; Association of Computer Machinery: New York, NY, USA, 2016; pp. 785–794. [Google Scholar]
- Breiman, L. Random Forests. Mach. Learn.
**2001**, 45, 5–32. [Google Scholar] [CrossRef][Green Version] - Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Dubourg, V.; et al. Scikit-Learn: Machine Learning in {P}ython. J. Mach. Learn. Res.
**2011**, 12, 2825–2830. [Google Scholar] - Saeys, W.; Nguyen Do Trong, N.; Van Beers, R.; Nicolaï, B.M. Multivariate Calibration of Spectroscopic Sensors for Postharvest Quality Evaluation: A Review. Postharvest Biol. Technol.
**2019**, 158, 110981. [Google Scholar] [CrossRef] - Kemps, B.J.; Saeys, W.; Mertens, K.; Darius, P.; de Baerdemaeker, J.G.; de Ketelaere, B. The Importance of Choosing the Right Validation Strategy in Inverse Modelling. J. Near Infrared Spectrosc.
**2010**, 18, 231–237. [Google Scholar] [CrossRef] - Luo, S.; Jiang, X.; He, Y.; Li, J.; Jiao, W.; Zhang, S.; Xu, F.; Han, Z.; Sun, J.; Yang, J.; et al. Multi-Dimensional Variables and Feature Parameter Selection for Aboveground Biomass Estimation of Potato Based on UAV Multispectral Imagery. Front. Plant Sci.
**2022**, 13, 2673. [Google Scholar] [CrossRef] - Bendig, J.; Yu, K.; Aasen, H.; Bolten, A.; Bennertz, S.; Broscheit, J.; Gnyp, M.L.; Bareth, G. Combining UAV-Based Plant Height from Crop Surface Models, Visible, and near Infrared Vegetation Indices for Biomass Monitoring in Barley. Int. J. Appl. Earth Obs. Geoinf.
**2015**, 39, 79–87. [Google Scholar] [CrossRef] - Pranga, J.; Borra-Serrano, I.; Aper, J.; de Swaef, T.; Ghesquiere, A.; Quataert, P.; Roldán-Ruiz, I.; Janssens, I.A.; Ruysschaert, G.; Lootens, P. Improving Accuracy of Herbage Yield Predictions in Perennial Ryegrass with UAV-Based Structural and Spectral Data Fusion and Machine Learning. Remote Sens.
**2021**, 13, 3459. [Google Scholar] [CrossRef] - Viljanen, N.; Honkavaara, E.; Näsi, R.; Hakala, T.; Niemeläinen, O.; Kaivosoja, J. A Novel Machine Learning Method for Estimating Biomass of Grass Swards Using a Photogrammetric Canopy Height Model, Images and Vegetation Indices Captured by a Drone. Agriculture
**2018**, 8, 70. [Google Scholar] [CrossRef][Green Version] - Näsi, R.; Viljanen, N.; Kaivosoja, J.; Alhonoja, K.; Hakala, T.; Markelin, L.; Honkavaara, E. Estimating Biomass and Nitrogen Amount of Barley and Grass Using UAV and Aircraft Based Spectral and Photogrammetric 3D Features. Remote Sens.
**2018**, 10, 1082. [Google Scholar] [CrossRef][Green Version] - Štroner, M.; Urban, R.; Seidl, J.; Reindl, T.; Brouček, J. Photogrammetry Using UAV-Mounted GNSS RTK: Georeferencing Strategies without GCPs. Remote Sens.
**2021**, 13, 1336. [Google Scholar] [CrossRef]

**Figure 1.**The location of the leek fields at the research centres (yellow) and farmers (green) during the 2019 (triangles) and 2020 (squares) experiments.

**Figure 3.**Evolution of the NDVI values acquired at the field of PSKW during the 2019 growing season. The cyan rectangles indicate the complete experimental plot while the red rectangle indicates the area used to extract the multispectral data.

**Figure 4.**The evolution of (

**a**) the DBM (g plant

^{−1}) and (

**b**) the N-uptake (gN plant

^{−1}) across all experimental plots in 2019 (blue) and 2020 (orange) in function of the days after planting. The solid line is a lowess smoother.

**Figure 5.**$RMS{E}_{ct}$ values for all the SLR models for (

**a**) DBM (g plant

^{−1}) and (

**b**) N-uptake (gN plant

^{−1}) ordered from the smallest to largest.

**Figure 6.**Scatter plots of predicted versus observed values for (

**a**) leek DBM using the Lasso model and (

**b**) leek N-uptake using the SLR model based on the median RWDRVI. The red dotted line shows the lowess smoother for the predictions.

**Figure 7.**The predicted dbm (

**a**) and N-uptake (

**b**) in function of their observed values at the research centres (orange dots) and farmer fields (blue crosses) from the experiments in 2020.

**Table 1.**Description of the fields that were monitored during the experiments, the varieties that were planted, fertilizers used, the number of observations and flights for both the plant nitrogen uptake (N-uptake) and dry biomass (DBM) and the days after planting (DAP) on which the flights were executed.

Season | Type | Location | Area (ha) | Variety | Fertilizer | Plant Density (ha^{−1}) | n_{plots} | n_{flights} | n_{obs} | DAP |
---|---|---|---|---|---|---|---|---|---|---|

2019 | Research Centre | ILVO ^{1} | 0.89 | Pluston | CAN | 153 846 | 12 | 3 | 40 | 41, 99, 125 |

Inagro ^{2} | 0.53 | Pluston | CAN | 153 846 | 24 | 5 | 120 | 41, 101, 132, 176, 216 | ||

PCG ^{2} | 0.14 | Pluston | CAN | 153 846 | 12 | 4 | 48 | 3, 41, 100, 129 | ||

PSKW ^{1} | 0.09 | Pluston | CAN | 178 571 | 8 | 5 | 40 | 6, 41, 100, 126, 168 | ||

2020 | Research Centre | ILVO ^{1} | 0.48 | Pluston | CAN | 153 846 | 16 | 4 | 64 | 44, 83, 136, 251 |

Inagro ^{2} | 0.32 | Pluston | CAN | 153 846 | 16 | 4 | 64 | 49, 91, 135, 210 | ||

PCG ^{2} | 0.18 | Pluston | CAN | 153 846 | 4 | 3 | 12 | 42, 84, 129 | ||

PSKW ^{1} | 0.28 | Pluston | CAN | 178 571 | 16 | 4 (1) * | 64 (16) * | 36, 77, 105, 149 * | ||

Farmer | F1 ^{2} | 3.51 | Vitaton | PS + Urean | 185 185 | 3 | 3 | 9 | 43, 124, 257 | |

F2 ^{1} | 2.97 | Chiefton | CM + CAN | 185 185 | 2 | 2 | 4 | 43, 134 | ||

Pluston | CM + CAN | 185 185 | 1 | 2 | 2 | 38, 129 | ||||

F3 ^{2} | 1.45 | Pluston | CS + CAN | 161 943 | 3 | 3 | 9 | 36, 118, 183 | ||

F4 ^{2} | 3.09 | Krypton | N.A. ^{§} | 161 943 | 2 | 2 | 4 | 54, 243 | ||

F5 ^{1} | 2.14 | Harston | PS + CAN | 170 940 | 2 | 3 (1) * | 6 (2) * | 44, 100, 266 * | ||

F6 ^{2} | 1.88 | Vitaton | CS + CAN | 151 515 | 2 | 1 | 2 | 204 | ||

F7 ^{2} | 1.21 | Pluston | PS + CAN | 175 439 | 1 | 3 | 3 | 40, 122, 176 | ||

Poulton | PS + CAN | 175 439 | 2 | 3 | 6 | 56, 138, 187 | ||||

F8 ^{2} | 0.79 | Harston | CM + APP | 208 333 | 2 | 2 | 4 | 47, 263 | ||

F9 ^{1} | 1.95 | Pluston | Novatec | 227 273 | 2 | 3 (1) * | 6 (2) * | 37, 106, 150 * | ||

F10 ^{1} | 0.66 | Poulton | RM + CAN | 170 940 | 2 | 2 (0) * | 4 (0) * | 43, 134 | ||

Total 2019 | 56 | 17 | 248 | |||||||

Total 2020 | 76 | 44 (32) * | 263 (203) * | |||||||

Total | 132 | 61 | 511 (451) * |

^{1,2}These UAV flights were executed by certified pilots with a DJI M600

^{1}or a DJI M200

^{2}. * The number of flights and observations that were matched with N-uptake observations deviates from the number of flights for DBM observations and is indicated between brackets. The flight moments that correspond to those flights is also indicated with *. The different mineral fertilizers used were calcium ammonia nitrate (CAN), Urean, Novatec, and ammonia polyphosphate (APP), while the used organic fertilizers were pig slurry (PS), cattle slurry (CS) or cattle manure (CM).

^{§}The farmer did not provide information on the type of fertilizer used.

VI | Formula | References |
---|---|---|

NDVI | (nir − red)/(nir + red) | [29] |

NDRE | (nir − rededge)/(nir + rededge) | [30] |

GNDVI | (nir − green)/(nir + green) | [31] |

BNDVI | (nir − blue)/(nir + blue) | [32] |

RWDRVI | (0.1 × nir − red)/(0.1× nir + red) | [33] |

BWDRVI | (0.1 × nir − blue)/(0.1× nir + blue) | [33] |

MTCI | (nir − rededge)/(rededge − red) | [34] |

MCARI | (rededge − red) − 0.2(rededge − green)(rededge/red) | [35] |

CCCI | [(nir + red)(nir − rededge)]/[(nir − red)(nir + rededge)] | [36] |

CI-green | nir/green − 1 | [37] |

CI-rededge | nir/rededge − 1 | [37] |

SAVI | [(1 + 0.5) × (nir − rededge)]/(nir + red + 0.5) | [38] |

**Table 3.**Summary of the model performances based on ${\overline{RMSE}}_{train}$ across all LOF-CV iterations, the RMSE

_{cv}, Bias

_{ct}, and R

^{2}for DBM (g plant

^{−1}). The SLR model performance is based on the median RWDRVI model.

SLR | Lasso | PLSR | SVR | RFR | xGBR | |
---|---|---|---|---|---|---|

${\overline{RMSE}}_{train}$ (g plant^{−1}) | 8.11 | 5.00 | 5.26 | 3.59 | 3.30 | 1.97 |

${\overline{RMSE}}_{cv}$ (g plant^{−1}) | 6.34 | 7.22 | 7.58 | 11.15 | 6.37 | 6.31 |

RMSE_{ct} (g plant^{−1}) | 9.08 | 6.60 | 7.90 | 10.92 | 6.35 | 6.72 |

rRMSE_{ct} (%) | 42.97 | 31.21 | 37.36 | 51.67 | 30.03 | 31.80 |

Bias_{ct} (g plant^{−1}) | −0.02 | 0.59 | 0.70 | −0.07 | −0.43 | −0.89 |

R^{2} | 0.81 | 0.90 | 0.87 | 0.73 | 0.91 | 0.90 |

**Table 4.**Summary of the model performance based on ${\overline{RMSE}}_{train}$ across all LOF-CV iterations, the RMSE

_{cv}, Bias

_{ct}, and R

^{2}for N-uptake (gN plant

^{−1}). The SLR model performance is based on the median RWDRVI model.

SLR | Lasso | PLSR | SVR | RFR | xGBR | |
---|---|---|---|---|---|---|

${\overline{RMSE}}_{train}$ (g plant^{−1}) | 0.21 | 0.20 | 0.16 | 0.08 | 0.11 | 0.15 |

${\overline{RMSE}}_{cv}$ (g plant^{−1}) | 0.20 | 0.25 | 0.23 | 0.32 | 0.19 | 0.18 |

RMSE_{ct} (g plant^{−1}) | 0.22 | 0.29 | 0.26 | 0.30 | 0.21 | 0.23 |

rRMSE_{ct} (%) | 39.08 | 51.35 | 44.89 | 52.49 | 36.78 | 40.11 |

Bias_{ct} (g plant^{−1}) | 0.00 | 0.05 | 0.01 | 0.14 | 0.03 | −0.05 |

R^{2} | 0.85 | 0.75 | 0.82 | 0.78 | 0.87 | 0.87 |

**Table 5.**Summary of the Lasso model to predict Pluston DBM and SLR model to predict Pluston N-uptake on the test data acquired during the 2020 experiments.

Field Type | All | Research Centre | Farmer | |
---|---|---|---|---|

Variety | Pluston | Pluston | Pluston | |

DBM | ||||

n_{obs} | 224 | 204 | 20 | |

RMSEP | 8.50 | 8.37 | 9.79 | |

rRMSEP | 49.98 | 49.96 | 50.62 | |

Bias | 1.28 | 1.10 | 3.11 | |

R^{2} | 0.77 | 0.77 | 0.78 | |

N-uptake | ||||

n_{obs} | 172 | 156 | 16 | |

RMSEP | 0.27 | 0.28 | 0.24 | |

rRMSEP | 69.77 | 72.76 | 49.51 | |

Bias | 0.03 | 0.02 | 0.09 | |

R^{2} | 0.68 | 0.67 | 0.82 |

**Table 6.**Summary of the prediction accuracy for leek DBM and N-uptake for leek varieties that were not included in the training dataset.

Variety | Chiefton | Harston | Krypton | Poulton | Vitaton | |
---|---|---|---|---|---|---|

DBM | ||||||

n_{obs} | 4 | 10 | 4 | 10 | 11 | |

RMSEP | 6.19 | 21.68 | 9.20 | 13.32 | 11.59 | |

Bias | 1.03 | 3.89 | 3.94 | 9.83 | 0.23 | |

R^{2} | 1.00 | 0.59 | 0.86 | 0.71 | 0.70 | |

N-uptake | ||||||

n_{obs} | 4 | 6 | 4 | 6 | 11 | |

RMSEP | 0.27 | 0.26 | 0.26 | 0.30 | 0.44 | |

Bias | 0.24 | −0.01 | 0.04 | 0.23 | 0.29 | |

R^{2} | 0.96 | 0.73 | 0.77 | 0.96 | 0.65 |

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

© 2022 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**

Haumont, J.; Lootens, P.; Cool, S.; Van Beek, J.; Raymaekers, D.; Ampe, E.; De Cuypere, T.; Bes, O.; Bodyn, J.; Saeys, W.
Multispectral UAV-Based Monitoring of Leek Dry-Biomass and Nitrogen Uptake across Multiple Sites and Growing Seasons. *Remote Sens.* **2022**, *14*, 6211.
https://doi.org/10.3390/rs14246211

**AMA Style**

Haumont J, Lootens P, Cool S, Van Beek J, Raymaekers D, Ampe E, De Cuypere T, Bes O, Bodyn J, Saeys W.
Multispectral UAV-Based Monitoring of Leek Dry-Biomass and Nitrogen Uptake across Multiple Sites and Growing Seasons. *Remote Sensing*. 2022; 14(24):6211.
https://doi.org/10.3390/rs14246211

**Chicago/Turabian Style**

Haumont, Jérémie, Peter Lootens, Simon Cool, Jonathan Van Beek, Dries Raymaekers, Eva Ampe, Tim De Cuypere, Onno Bes, Jonas Bodyn, and Wouter Saeys.
2022. "Multispectral UAV-Based Monitoring of Leek Dry-Biomass and Nitrogen Uptake across Multiple Sites and Growing Seasons" *Remote Sensing* 14, no. 24: 6211.
https://doi.org/10.3390/rs14246211