Fusion of Ultrasonic and Spectral Sensor Data for Improving the Estimation of Biomass in Grasslands with Heterogeneous Sward Structure

Evaluating UAV-Based Remote Sensing for Hay Yield Estimation

(1) Background: Yield-monitoring systems are widely used in grain crops but are less advanced for hay and forage. Current commercial systems are generally limited to weighing individual bales, limiting the spatial resolution of maps of hay yield. This study evaluated an Uncrewed Aerial Vehicle (UAV)-based imaging system to estimate hay yield. (2) Methods: Data were collected from three 0.4 ha plots and a 35 ha hay field of red clover and timothy grass in September 2020. A multispectral camera on the UAV captured images at 30 m (20 mm pixel-1) and 50 m (35 mm pixel-1) heights. Eleven Vegetation Indices (VIs) and five texture features were calculated from the images to estimate biomass yield. Multivariate regression models (VIs and texture features vs. biomass) were evaluated. (3) Results: Model R2 values ranged from 0.31 to 0.68. (4) Conclusions: Despite strong correlations between standard VIs and biomass, challenges such as variable image resolution and clarity affected accuracy. Further research is needed before UAV-based yield estimation can provide accurate, high-resolution hay yield maps.

Utilizing Spectral, Structural and Textural Features for Estimating Oat Above-Ground Biomass Using UAV-Based Multispectral Data and Machine Learning

Accurate and timely monitoring of biomass in breeding nurseries is essential for evaluating plant performance and selecting superior genotypes. Traditional methods for phenotyping above-ground biomass in field conditions requires significant time, cost, and labor. Unmanned Aerial Vehicles (UAVs) offer a rapid and non-destructive approach for phenotyping multiple field plots at a low cost. While Vegetation Indices (VIs) extracted from remote sensing imagery have been widely employed for biomass estimation, they mainly capture spectral information and disregard the 3D canopy structure and spatial pixel relationships. Addressing these limitations, this study, conducted in 2020 and 2021, aimed to explore the potential of integrating UAV multispectral imagery-derived canopy spectral, structural, and textural features with machine learning algorithms for accurate oat biomass estimation. Six oat genotypes planted at two seeding rates were evaluated in two South Dakota locations at multiple growth stages. Plot-level canopy spectral, structural, and textural features were extracted from the multispectral imagery and used as input variables for three machine learning models: Partial Least Squares Regression (PLSR), Support Vector Regression (SVR), and Random Forest Regression (RFR). The results showed that (1) in addition to canopy spectral features, canopy structural and textural features are also important indicators for oat biomass estimation; (2) combining spectral, structural, and textural features significantly improved biomass estimation accuracy over using a single feature type; (3) machine learning algorithms showed good predictive ability with slightly better estimation accuracy shown by RFR (R2 = 0.926 and relative root mean square error (RMSE%) = 15.97%). This study demonstrated the benefits of UAV imagery-based multi-feature fusion using machine learning for above-ground biomass estimation in oat breeding nurseries, holding promise for enhancing the efficiency of oat breeding through UAV-based phenotyping and crop management practices.

Within and combined season prediction models for perennial ryegrass biomass yield using ground- and air-based sensor data

Across-season biomass assessment is crucial in the cultivar selection process to accurately evaluate the yield performance of lines under different growing conditions. However, it has been difficult to have an accurate, reliable, and repeated fresh biomass (FM) estimation of large populations of plants in the field without destructive harvesting, which incurs significant labor and operation costs. Sensor-based phenotyping platforms have advanced in the data collection of structural and vegetative information of plants, but the developed prediction models are still limited by low correlations at different growth stages and seasons. In this study, our objective was to develop and validate the global prediction models for across-season harvested fresh biomass (FM) yield based on the ground- and air-based sensor data including ground-based LiDAR, ground-based ultrasonic, and air-based multispectral camera to extract LiDAR plant volume (LV), LiDAR point density (LV_Den), height, and Normalized Difference Vegetative Index (NDVI). The study was conducted in a row-plot field trial with 480 rows (3 rows in a plot per cultivar) throughout the whole 2020 growing season up to the reproductive stage. We evaluated the performance of each plant parameter, their relationship, and the best subset prediction models using statistical stepwise selection at the row and plot levels through the seasonal and combined seasonal datasets. The best performing model: F M ~ L V ∗ L V _ D e n ∗ N D V I had a determination of coefficient R ² of at least 0.9 in vegetative stages and 0.8 in the reproductive stage. Similar results can be achieved in a simpler model with just two LiDAR variables- F M ~ L V ∗ L V _ D e n . In addition, LV and LV_Den showed a robust correlation with FM on their own over seasons and growth stages, while NDVI only performed well in some seasons. The simpler model based on only LiDAR data can be widely applied over season without the need of additional sensor data and may thus make the in-field across-season biomass assessment more feasible and practical for fast and cost-effective development of higher biomass yield cultivars.

Data Fusion in Agriculture: Resolving Ambiguities and Closing Data Gaps

Acquiring useful data from agricultural areas has always been somewhat of a challenge, as these are often expansive, remote, and vulnerable to weather events. Despite these challenges, as technologies evolve and prices drop, a surge of new data are being collected. Although a wealth of data are being collected at different scales (i.e., proximal, aerial, satellite, ancillary data), this has been geographically unequal, causing certain areas to be virtually devoid of useful data to help face their specific challenges. However, even in areas with available resources and good infrastructure, data and knowledge gaps are still prevalent, because agricultural environments are mostly uncontrolled and there are vast numbers of factors that need to be taken into account and properly measured for a full characterization of a given area. As a result, data from a single sensor type are frequently unable to provide unambiguous answers, even with very effective algorithms, and even if the problem at hand is well defined and limited in scope. Fusing the information contained in different sensors and in data from different types is one possible solution that has been explored for some decades. The idea behind data fusion involves exploring complementarities and synergies of different kinds of data in order to extract more reliable and useful information about the areas being analyzed. While some success has been achieved, there are still many challenges that prevent a more widespread adoption of this type of approach. This is particularly true for the highly complex environments found in agricultural areas. In this article, we provide a comprehensive overview on the data fusion applied to agricultural problems; we present the main successes, highlight the main challenges that remain, and suggest possible directions for future research.

Potentials and Limitations of WorldView-3 Data for the Detection of Invasive Lupinus polyphyllus Lindl. in Semi-Natural Grasslands

Semi-natural grasslands contribute highly to biodiversity and other ecosystem services, but they are at risk by the spread of invasive plant species, which alter their habitat structure. Large area grassland monitoring can be a powerful tool to manage invaded ecosystems. Therefore, WorldView-3 multispectral sensor data was utilized to train multiple machine learning algorithms in an automatic machine learning workflow called ‘H2O AutoML’ to detect L. polyphyllus in a nature protection grassland ecosystem. Different degree of L. polyphyllus cover was collected on 3 × 3 m2 reference plots, and multispectral bands, indices, and texture features were used in a feature selection process to identify the most promising classification model and machine learning algorithm based on mean per class error, log loss, and AUC metrics. The best performance was achieved with a binary classification of lupin-free vs. fully invaded 3 × 3 m2 plot classification with a set of 7 features out of 763. The findings reveal that L. polyphyllus detection from WorldView-3 sensor data is limited to large dominant spots and not recommendable for lower plant coverage, especially single plant detection. Further research is needed to clarify if different phenological stages of L. polyphyllus as well as time series increase classification performance.

UAV Data as an Alternative to Field Sampling to Monitor Vineyards Using Machine Learning Based on UAV/Sentinel-2 Data Fusion

Pests and diseases affect the yield and quality of grapes directly and engender noteworthy economic losses. Diagnosing “lesions” on vines as soon as possible and dynamically monitoring symptoms caused by pests and diseases at a larger scale are essential to pest control. This study has appraised the capabilities of high-resolution unmanned aerial vehicle (UAV) data as an alternative to manual field sampling to obtain sampling canopy sets and to supplement satellite-based monitoring using machine learning models including partial least squared regression (PLSR), support vector regression (SVR), random forest regression (RFR), and extreme learning regression (ELR) with a new activation function. UAV data were acquired from two flights in Turpan to determine disease severity (DS) and disease incidence (DI) and compared with field visual assessments. The UAV-derived canopy structure including canopy height (CH) and vegetation fraction cover (VFC), as well as satellite-based spectral features calculated from Sentinel-2A/B data were analyzed to evaluate the potential of UAV data to replace manual sampling data and predict DI. It was found that SVR slightly outperformed the other methods with a root mean square error (RMSE) of 1.89%. Moreover, the combination of canopy structure (CS) and vegetation index (VIs) improved prediction accuracy compared with single-type features (RMSEcs of 2.86% and RMSEVIs of 1.93%). This study tested the ability of UAV sampling to replace manual sampling on a large scale and introduced opportunities and challenges of fusing different features to monitor vineyards using machine learning. Within this framework, disease incidence can be estimated efficiently and accurately for larger area monitoring operation.

Development of Machine Learning Models to Predict Compressed Sward Height in Walloon Pastures Based on Sentinel-1, Sentinel-2 and Meteorological Data Using Multiple Data Transformations

Accurate information about the available standing biomass on pastures is critical for the adequate management of grazing and its promotion to farmers. In this paper, machine learning models are developed to predict available biomass expressed as compressed sward height (CSH) from readily accessible meteorological, optical (Sentinel-2) and radar satellite data (Sentinel-1). This study assumed that combining heterogeneous data sources, data transformations and machine learning methods would improve the robustness and the accuracy of the developed models. A total of 72,795 records of CSH with a spatial positioning, collected in 2018 and 2019, were used and aggregated according to a pixel-like pattern. The resulting dataset was split into a training one with 11,625 pixellated records and an independent validation one with 4952 pixellated records. The models were trained with a 19-fold cross-validation. A wide range of performances was observed (with mean root mean square error (RMSE) of cross-validation ranging from 22.84 mm of CSH to infinite-like values), and the four best-performing models were a cubist, a glmnet, a neural network and a random forest. These models had an RMSE of independent validation lower than 20 mm of CSH at the pixel-level. To simulate the behavior of the model in a decision support system, performances at the paddock level were also studied. These were computed according to two scenarios: either the predictions were made at a sub-parcel level and then aggregated, or the data were aggregated at the parcel level and the predictions were made for these aggregated data. The results obtained in this study were more accurate than those found in the literature concerning pasture budgeting and grassland biomass evaluation. The training of the 124 models resulting from the described framework was part of the realization of a decision support system to help farmers in their daily decision making.

Estimation of forage biomass and vegetation cover in grasslands using UAV imagery

Grasslands are among the most widespread ecosystems on Earth and among the most degraded. Their characterization and monitoring are generally based on field measurements, which are incomplete spatially and temporally. The recent advent of unmanned aerial vehicles (UAV) provides data at unprecedented spatial and temporal resolutions. This study aims to test and compare three approaches based on multispectral imagery acquired by UAV to estimate forage biomass or vegetation cover in grasslands. The study site is composed of 30 pasture plots (25 × 50 m), 5 bare soil plots (25 x 50), and 6 control plots (5 × 5 m) on a 14-ha field maintained at various biomass levels by grazing rotations and clipping over a complete growing season. A total of 14 flights were performed. A first approach based on structure from motion was used to generate a volumetric-based biomass estimation model (R2 of 0.93 and 0.94 for fresh biomass [FM] and dry biomass [DM], respectively). This approach is not very sensitive to low vegetation levels but is accurate for FM estimation greater than 0.5 kg/m2 (0.1 kg DM/m2). The Green Normalized Difference Vegetation Index (GNDVI) was selected to develop two additional approaches. One is based on a regression biomass prediction model (R2 of 0.80 and 0.66 for FM and DM, respectively) and leads to an accurate estimation at levels of FM lower than 3 kg/m2 (0.6 kg DM/m2). The other approach is based on a classification of vegetation cover from clustering of GNDVI values in four classes. This approach is more qualitative than the other ones but more robust and generalizable. These three approaches are relatively simple to use and applicable in an operational context. They are also complementary and can be adapted to specific applications in grassland characterization.

Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB Imagery

Monitoring biomass of forages in experimental plots and livestock farms is a time-consuming, expensive, and biased task. Thus, non-destructive, accurate, precise, and quick phenotyping strategies for biomass yield are needed. To promote high-throughput phenotyping in forages, we propose and evaluate the use of deep learning-based methods and UAV (Unmanned Aerial Vehicle)-based RGB images to estimate the value of biomass yield by different genotypes of the forage grass species Panicum maximum Jacq. Experiments were conducted in the Brazilian Cerrado with 110 genotypes with three replications, totaling 330 plots. Two regression models based on Convolutional Neural Networks (CNNs) named AlexNet and ResNet18 were evaluated, and compared to VGGNet-adopted in previous work in the same thematic for other grass species. The predictions returned by the models reached a correlation of 0.88 and a mean absolute error of 12.98% using AlexNet considering pre-training and data augmentation. This proposal may contribute to forage biomass estimation in breeding populations and livestock areas, as well as to reduce the labor in the field.

Crop Monitoring Using Satellite/UAV Data Fusion and Machine Learning

Non-destructive crop monitoring over large areas with high efficiency is of great significance in precision agriculture and plant phenotyping, as well as decision making with regards to grain policy and food security. The goal of this research was to assess the potential of combining canopy spectral information with canopy structure features for crop monitoring using satellite/unmanned aerial vehicle (UAV) data fusion and machine learning. Worldview-2/3 satellite data were tasked synchronized with high-resolution RGB image collection using an inexpensive unmanned aerial vehicle (UAV) at a heterogeneous soybean (Glycine max (L.) Merr.) field. Canopy spectral information (i.e., vegetation indices) was extracted from Worldview-2/3 data, and canopy structure information (i.e., canopy height and canopy cover) was derived from UAV RGB imagery. Canopy spectral and structure information and their combination were used to predict soybean leaf area index (LAI), aboveground biomass (AGB), and leaf nitrogen concentration (N) using partial least squares regression (PLSR), random forest regression (RFR), support vector regression (SVR), and extreme learning regression (ELR) with a newly proposed activation function. The results revealed that: (1) UAV imagery-derived high-resolution and detailed canopy structure features, canopy height, and canopy coverage were significant indicators for crop growth monitoring, (2) integration of satellite imagery-based rich canopy spectral information with UAV-derived canopy structural features using machine learning improved soybean AGB, LAI, and leaf N estimation on using satellite or UAV data alone, (3) adding canopy structure information to spectral features reduced background soil effect and asymptotic saturation issue to some extent and led to better model performance, (4) the ELR model with the newly proposed activated function slightly outperformed PLSR, RFR, and SVR in the prediction of AGB and LAI, while RFR provided the best result for N estimation. This study introduced opportunities and limitations of satellite/UAV data fusion using machine learning in the context of crop monitoring.

Prediction of Biomass and N Fixation of Legume-Grass Mixtures Using Sensor Fusion

European farmers and especially organic farmers rely on legume-grass mixtures in their crop rotation as an organic nitrogen (N) source, as legumes can fix atmospheric N, which is the most important element for plant growth. Furthermore, legume-grass serves as valuable fodder for livestock and biogas plants. Therefore, information about aboveground biomass and N fixation (NFix) is crucial for efficient farm management decisions on the field level. Remote sensing, as a non-destructive and fast technique, provides different methods to quantify plant trait parameters. In our study, high-density point clouds, derived from terrestrial laser scanning (TLS), in combination with unmanned aerial vehicle-based multispectral (MS) data, were collected to receive information about three plant trait parameters (fresh and dry matter, nitrogen fixation) in two legume-grass mixtures. Several crop surface height metrics based on TLS and vegetation indices based on the four MS bands (green, red, red edge, and near-infrared) were calculated. Furthermore, eight texture features based on mean crop surface height and the four MS bands were generated to measure horizontal spatial heterogeneity. The aim of this multi-temporal study over two vegetation periods was to create estimation models based on biomass and N fixation for two legume-grass mixtures by sensor fusion, a combination of both sensors. To represent conditions in practical farming, e.g., the varying proportion of legumes, the experiment included pure stands of legume and grass of the mixtures. Sensor fusion of TLS and MS data was found to provide better estimates of biomass and N _Fix than separate data analysis. The study shows the important role of texture based on MS and point cloud data, which contributed greatly to the estimation model generation. The applied approach offers an interesting method for improvements in precision agriculture.

Ultrasonic Arrays for Remote Sensing of Pasture Biomass

The profitability of agricultural industries that utilise pasture can be strongly affected by the ability to accurately measure pasture biomass. Pasture height measurement is one technique that has been used to estimate pasture biomass. However, pasture height measurement errors can occur if the sensor is mounted to a farm vehicle that experiences tilting or bouncing. This work describes the development of novel low ultrasonic frequency arrays for pasture biomass estimation. Rather than just measuring the distance to the top of the pasture, as previous ultrasonic studies have done, this hardware is designed to also allow ultrasonic measurements to be made vertically through the pasture to the ground. The hardware was mounted to a farm bike driving over pasture at speeds of up to 20 km/h. The analysed results show the ability of the hardware to measure the ground location through the grass. This allowed pasture height measurement to be independent of tilting and bouncing of the farm vehicle, leading to 20 to 25% improvement in the R 2 value obtained for biomass estimation compared with the traditional technique. This corresponded to a reduction in root mean squared error of predicted biomass from about 350 to 270 kg/ha, where the average biomass of the pasture was 1915 kg/ha.

Ultrasonic Proximal Sensing of Pasture Biomass

The optimization of pasture food value, known as ‘biomass’, is crucial in the management of the farming of grazing animals and in improving food production for the future. Optical sensing methods, particularly from satellite platforms, provide relatively inexpensive and frequently updated wide-area coverage for monitoring biomass and other forage properties. However, there are also benefits from direct or proximal sensing methods for higher accuracy, more immediate results, and for continuous updates when cloud cover precludes satellite measurements. Direct measurement, by cutting and weighing the pasture, is destructive, and may not give results representative of a larger area of pasture. Proximal sensing methods may also suffer from sampling small areas, and can be generally inaccurate. A new proximal methodology is described here, in which low-frequency ultrasound is used as a sonar to obtain a measure of the vertical variation of the pasture density between the top of the pasture and the ground and to relate this to biomass. The instrument is designed to operate from a farm vehicle moving at up to 20 km h−1, thus allowing a farmer to obtain wide coverage in the normal course of farm operations. This is the only method providing detailed biomass profile information from throughout the entire pasture canopy. An essential feature is the identification of features from the ultrasonic reflectance, which can be related sensibly to biomass, thereby generating a physically-based regression model. The result is significantly improved estimation of pasture biomass, in comparison with other proximal methods. Comparing remotely sensed biomass to the biomass measured via cutting and weighing gives coefficients of determination, R2, in the range of 0.7 to 0.8 for a range of pastures and when operating the farm vehicle at speeds of up to 20 km h−1.

Biomass Prediction of Heterogeneous Temperate Grasslands Using an SfM Approach Based on UAV Imaging

An early and precise yield estimation in intensive managed grassland is mandatory for economic management decisions. RGB (red, green, blue) cameras attached on an unmanned aerial vehicle (UAV) represent a promising non-destructive technology for the assessment of crop traits especially in large and remote areas. Photogrammetric structure from motion (SfM) processing of the UAV-based images into point clouds can be used to generate 3D spatial information about the canopy height (CH). The aim of this study was the development of prediction models for dry matter yield (DMY) in temperate grassland based on CH data generated by UAV RGB imaging over a whole growing season including four cuts. The multi-temporal study compared the remote sensing technique with two conventional methods, i.e., destructive biomass sampling and ruler height measurements in two legume-grass mixtures with red clover (Trifolium pratense L.) and lucerne (Medicago sativa L.) in combination with Italian ryegrass (Lolium multiflorum Lam.). To cover the full range of legume contribution occurring in a practical grassland, pure stands of legumes and grasses contained in each mixture were also investigated. The results showed, that yield prediction by SfM-based UAV RGB imaging provided similar accuracies across all treatments (R2 = 0.59–0.81) as the ruler height measurements (R2 = 0.58–0.78). Furthermore, results of yield prediction by UAV RGB imaging demonstrated an improved robustness when an increased CH variability occurred due to extreme weather conditions. It became apparent that morphological characteristics of clover-based canopies (R2 = 0.75) allow a better remotely sensed prediction of total annual yield than for lucerne-grass mixtures (R2 = 0.64), and that these crop-specific models cannot be easily transferred to other grassland types.

Monitoring Seasonal Pasture Quality Degradation in the Mediterranean Montado Ecosystem: Proximal versus Remote Sensing

Montado is an agro-forestry system occupying a large surface in countries of the Mediterranean region. In this system, the natural dryland pasture is the principal source for animal feed in extensive grazing. The climatic seasonality associated with the inter-annual irregularity of precipitation greatly influences the development of pasture and its vegetative cycle. The end of spring is a critical period in terms of animal feed due to the notable reduction in the nutritive value of the plants. The objective of this work was to evaluate, through the correlation between pasture quality indexes (Pasture Quality Degradation Index, PQDI and Normalized Difference Vegetation Index, NDVI), two technological approaches for monitoring the evolution of the quality of a biodiverse pasture in the period of greatest vegetative development (between February and June). The technological approaches consisted of (i) proximal sensing (PS), with the use of an active optical sensor; and (ii) remote sensing (RS), using images captured by a Sentinel-2 satellite. The results of this study show strong and significant correlations between PQDI and NDVI (obtained by PS or RS). These two techniques (PS or RS) can, therefore, be used in a complementary way to identify and anticipate the food supplementation needs for animals and support farmers in decision making.

A Holistic Approach to the Evaluation of the Montado Ecosystem Using Proximal Sensors

The Montado is a silvo-pastoral system characterized by open canopy woodlands with natural or cultivated grassland in the undercover and grazing animals. The aims of this study were to present several proximal sensors with potential to monitor relevant variables in the complex montado ecosystem and demonstrate their application in a case study designed to evaluate the effect of trees on the pasture. This work uses data collected between March and June 2016, at peak of dryland pasture production under typical Mediterranean conditions, in twenty four sampling points, half under tree canopy (UTC) and half outside tree canopy (OTC). Correlations were established between pasture biomass and capacitance measured by a commercial probe and between pasture quality and normalized difference vegetation index (NDVI) measured by a commercial active optical sensor. The interest of altimetric and apparent soil electrical conductivity maps as the first step in the implementation of precision agriculture projects was demonstrated. The use of proximal sensors to monitor soil moisture content, pasture photosynthetically active radiation and temperature helped to explain the influence of trees on pasture productivity and quality. The significant and strong correlations obtained between capacitance and pasture biomass and between NDVI and pasture nutritive value (in terms of crude protein, CP and neutral detergent fibre, NDF) can make an important contribution to determination of key components of pasture productivity and quality and implementation of site-specific pasture management. Animal tracking demonstrated its potential to be an important tool for understanding the interaction between various factors and components that interrelate in the montado ecosystem and to support grazing management decisions.

Estimating Barley Biomass with Crop Surface Models from Oblique RGB Imagery

Non-destructive monitoring of crop development is of key interest for agronomy and crop breeding. Crop Surface Models (CSMs) representing the absolute height of the plant canopy are a tool for this. In this study, fresh and dry barley biomass per plot are estimated from CSM-derived plot-wise plant heights. The CSMs are generated in a semi-automated manner using Structure-from-Motion (SfM)/Multi-View-Stereo (MVS) software from oblique stereo RGB images. The images were acquired automatedly from consumer grade smart cameras mounted at an elevated position on a lifting hoist. Fresh and dry biomass were measured destructively at four dates each in 2014 and 2015. We used exponential and simple linear regression based on different calibration/validation splits. Coefficients of determination R 2 between 0.55 and 0.79 and root mean square errors (RMSE) between 97 and 234 g/m2 are reached for the validation of predicted vs. observed dry biomass, while Willmott’s refined index of model performance d r ranges between 0.59 and 0.77. For fresh biomass, R 2 values between 0.34 and 0.61 are reached, with root mean square errors (RMSEs) between 312 and 785 g/m2 and d r between 0.39 and 0.66. We therefore established the possibility of using this novel low-cost system to estimate barley dry biomass over time.