Wavelength Selection Method Based on Partial Least Square from Hyperspectral Unmanned Aerial Vehicle Orthomosaic of Irrigated Olive Orchards

Estimating Community-Level Plant Functional Traits in a Species-Rich Alpine Meadow Using UAV Image Spectroscopy

Plant functional traits at the community level (plant community traits hereafter) are commonly used in trait-based ecology for the study of vegetation–environment relationships. Previous studies have shown that a variety of plant functional traits at the species or community level can be successfully retrieved by airborne or spaceborne imaging spectrometer in homogeneous, species-poor ecosystems. However, findings from these studies may not apply to heterogeneous, species-rich ecosystems. Here, we aim to determine whether unmanned aerial vehicle (UAV)-based hyperspectral imaging could adequately estimate plant community traits in a species-rich alpine meadow ecosystem on the Qinghai–Tibet Plateau. To achieve this, we compared the performance of four non-parametric regression models, i.e., partial least square regression (PLSR), the generic algorithm integrated with the PLSR (GA-PLSR), random forest (RF) and extreme gradient boosting (XGBoost) for the retrieval of 10 plant community traits using visible and near-infrared (450–950 nm) UAV hyperspectral imaging. Our results show that chlorophyll a, chlorophyll b, carotenoid content, starch content, specific leaf area and leaf thickness were estimated with good accuracies, with the highest R2 values between 0.64 (nRMSE = 0.16) and 0.83 (nRMSE = 0.11). Meanwhile, the estimation accuracies for nitrogen content, phosphorus content, plant height and leaf dry matter content were relatively low, with the highest R2 varying from 0.3 (nRMSE = 0.24) to 0.54 (nRMSE = 0.20). Among the four tested algorithms, the GA-PLSR produced the highest accuracy, followed by PLSR and XGBoost, and RF showed the poorest performance. Overall, our study demonstrates that UAV-based visible and near-infrared hyperspectral imaging has the potential to accurately estimate multiple plant community traits for the natural grassland ecosystem at a fine scale.

Hyperspectral Imagery Detects Water Deficit and Salinity Effects on Photosynthesis and Antioxidant Enzyme Activity of Three Greek Olive Varieties

The olive tree (Olea europaea L.) is one of the main crops of the Mediterranean region which suffers from drought and soil salinization. We assessed the photosynthetic rate, leaf water content and antioxidative enzyme activity (APX, GPX, SOD and CAT) of three Greek olive cultivars (‘Amfisis’, ‘Mastoidis’ and ‘Lefkolia Serron’) subjected to drought and salinity stresses. Hyperspectral reflectance data were acquired using an analytical spectral device (ASD) FieldSpec® 3 spectroradiometer, while principal component regression, partial least squares regression and linear discriminant analysis were used to estimate the relationship between spectral and physiological measurements. The photosynthetic rate and water content of stressed plants decreased, while enzyme activity had an increasing tendency. ‘Amfisis’ was more resistant to drought and salinity stress than ‘Mastoidis’ and ‘Lefkolia Serron’. The NDVI appeared to have the highest correlation with the photosynthetic rate, followed by the PRI. APX enzyme activity was the most highly correlated with the 1150–1370 nm range, with an additional peak at 1840 nm. CAT enzyme activity resulted in the highest correlation with the visible part of the spectrum with two peaks at 1480 nm and 1950 nm, while GPX enzyme activity appeared to have a strong correlation within all the available spectral ranges except for 670–1180 nm. Finally, SOD activity showed high correlation values within 1190–1850 nm. This is the first time the correlation of hyperspectral imagery with photosynthetic rate and antioxidant enzyme activities was determined, providing the background for high-throughput plant phenotyping through a drone with a hyperspectral camera. This progress would provide the possibility of early stress detection in large olive groves and assist farmers in decision making and optimizing crop management, health and productivity.

Investigating the potential of Sentinel-2 configuration to predict the quality of Mediterranean permanent grasslands in open woodlands

The assessment of pasture quality in permanent grasslands is essential for their conservation and management, as it can contribute to making real-time decisions for livestock management. In this study, we assessed the potential of Sentinel-2 configuration to predict forage quality in high diverse Mediterranean permanent grasslands of open woodlands. We evaluated the performance of Partial Least Squares Regression (PLS) models to predict crude protein (CP), neutral detergent fibre (NDF), acid detergent fibre (ADF) and enzyme digestibility of organic matter (EDOM) by using three different reflectance datasets: (i) laboratory measurements of reflectance of dry and ground pasture samples re-sampled to Sentinel-2 configuration (Spec-lab) (ii) field in-situ measurements of grasslands canopy reflectance resampled to Sentinel-2 configuration (Spec-field); (iii) and Bottom Of Atmosphere Sentinel-2 imagery. For the three reflectance datasets, the models to predict CP content showed moderate performance and predictive ability. Mean R²_test = 0.68 were obtained using Spec-lab data, mean R²_test decreased by 0.11 with Spec-field and by 0.18 when Sentinel-2 reflectance was used. Statistics for NDF showed worse predictions than those obtained for CP: predictions produced with Spec-lab showed mean R²_test = 0.64 and mean RPD_test = 1.73. The mean values of R²_test = 0.50 and RPD_test = 1.54 using Sentinel-2 BOA reflectance were marginally better than the values obtained with Spec-field (mean R²_test = 0.48, mean RPD_test = 1.43). For ADF and EDOM, only predictions made with Spec-lab produced acceptable results. Bands from the red-edge region, especially band 5, and the SWIR regions showed the highest contribution to estimating CP and NDF. Bands 2, blue and 4, red also seem to be important. The implementation of field spectroscopy in combination with Sentinel-2 imagery proved to be feasible to produce forage quality maps and to develop larger datasets. This study contributes to increasing knowledge of the potential and applicability of Sentinel-2 to predict the quality of Mediterranean permanent grasslands in open woodlands.

Analysis, Modeling and Multi-Spectral Sensing for the Predictive Management of Verticillium Wilt in Olive Groves

The intensification and expansion in the cultivation of olives have contributed to the significant spread of Verticillium wilt, which is the most important fungal problem affecting olive trees. Recent studies confirm that practices such as the use of innovative natural minerals (Zeoshell ZF1) and the application of beneficial microorganisms (Micosat F BS WP) restore health in infected trees. However, for their efficient implementation the above methodologies require the marking of trees in the early stages of infestation—a task that is impractical with traditional means (manual labor) but also very difficult, as early stages are difficult to perceive with the naked eye. In this paper, we present the results of the My Olive Grove Coach (MyOGC) project, which used multispectral imaging from unmanned aerial vehicles to develop an olive grove monitoring system based on the autonomous and automatic processing of the multispectral images using computer vision and machine learning techniques. The goal of the system is to monitor and assess the health of olive groves, help in the prediction of Verticillium wilt spread and implement a decision support system that guides the farmer/agronomist.

An Efficient Spectral Feature Extraction Framework for Hyperspectral Images

Extracting diverse spectral features from hyperspectral images has become a hot topic in recent years. However, these models are time consuming for training and test and suffer from a poor discriminative ability, resulting in low classification accuracy. In this paper, we design an effective feature extracting framework for the spectra of hyperspectral data. We construct a structured dictionary to encode spectral information and apply learning machine to map coding coefficients. To reduce training and testing time, the sparsity constraint is replaced by a block-diagonal constraint to accelerate the iteration, and an efficient extreme learning machine is employed to fit the spectral characteristics. To optimize the discriminative ability of our model, we first add spectral convolution to extract abundant spectral information. Then, we design shared constraints for subdictionaries so that the common features of subdictionaries can be expressed more effectively, and the discriminative and reconstructive ability of dictionary will be improved. The experimental results on diverse databases show that the proposed feature extraction framework can not only greatly reduce the training and testing time, but also lead to very competitive accuracy performance compared with deep learning models.

UAS-Remote Sensing Methods for Mapping, Monitoring and Modeling Crops

The advances in Unmanned Aerial Vehicle (UAV) platforms and on-board sensors in the past few years have greatly increased our ability to monitor and map crops. The ability to register images at ultra-high spatial resolution at any moment has made remote sensing techniques increasingly useful in crop management. These technologies have revolutionized the way in which remote sensing is applied in precision agriculture, allowing for decision-making in a matter of days instead of weeks. However, it is still necessary to continue research to improve and maximize the potential of UAV remote sensing in agriculture. This Special Issue of Remote Sensing includes different applications of UAV remote sensing for crop management, covering RGB, multispectral, hyperspectral and LIght Detection and Ranging (LiDAR) sensor applications on-board (UAVs). The papers reveal innovative techniques involving image analysis and cloud points. It should, however, be emphasized that this Special Issue is a small sample of UAV applications in agriculture and that there is much more to investigate.