Ultra-high-resolution UAV-imaging and supervised deep learning for accurate detection of Alternaria solani in potato fields.

Alternaria solani is the second most devastating foliar pathogen of potato crops worldwide, causing premature defoliation of the plants. This disease is currently prevented through the regular application of detrimental crop protection products and is guided by early warnings based on weather predictions and visual observations by farmers. To reduce the use of crop protection products, without additional production losses, it would be beneficial to be able to automatically detect Alternaria solani in potato fields. In recent years, the potential of deep learning in precision agriculture is receiving increasing research attention. Convolutional Neural Networks (CNNs) are currently the state of the art, but also come with challenges, especially regarding in-field robustness. This stems from the fact that they are often trained on datasets that are limited in size or have been recorded in controlled environments, not necessarily representative of real-world settings. We collected a dataset consisting of ultra-high-resolution modified RGB UAV-imagery of both symptomatic and non-symptomatic potato crops in the field during various years and disease stages to cover the great variability in agricultural data. We developed a convolutional neural network to perform in-field detection of Alternaria, defined as a binary classification problem. Our model achieves a similar accuracy as several state-of-the-art models for disease detection, but has a much lower inference time, which enhances its practical applicability. By using training data of three consecutive growing seasons (2019, 2020 and 2021) and test data of an independent fourth year (2022), an F1 score of 0.93 is achieved. Furthermore, we evaluate how different properties of the dataset such as its size and class imbalance impact the obtained accuracy.

Hyperspectral Classification of Cyperus esculentus Clones and Morphologically Similar Weeds.

Cyperus esculentus (yellow nutsedge) is one of the world's worst weeds as it can cause great damage to crops and crop production. To eradicate C. esculentus, early detection is key-a challenging task as it is often confused with other Cyperaceae and displays wide genetic variability. In this study, the objective was to classify C. esculentus clones and morphologically similar weeds. Hyperspectral reflectance between 500 and 800 nm was tested as a measure to discriminate between (I) C. esculentus and morphologically similar Cyperaceae weeds, and between (II) different clonal populations of C. esculentus using three classification models: random forest (RF), regularized logistic regression (RLR) and partial least squares-discriminant analysis (PLS-DA). RLR performed better than RF and PLS-DA, and was able to adequately classify the samples. The possibility of creating an affordable multispectral sensing tool, for precise in-field recognition of C. esculentus plants based on fewer spectral bands, was tested. Results of this study were compared against simulated results from a commercially available multispectral camera with four spectral bands. The model created with customized bands performed almost equally well as the original PLS-DA or RLR model, and much better than the model describing multispectral image data from a commercially available camera. These results open up the opportunity to develop a dedicated robust tool for C. esculentus recognition based on four spectral bands and an appropriate classification model.

Deep convolutional neural networks for image-based Convolvulus sepium detection in sugar beet fields.

BACKGROUND: Convolvulus sepium (hedge bindweed) detection in sugar beet fields remains a challenging problem due to variation in appearance of plants, illumination changes, foliage occlusions, and different growth stages under field conditions. Current approaches for weed and crop recognition, segmentation and detection rely predominantly on conventional machine-learning techniques that require a large set of hand-crafted features for modelling. These might fail to generalize over different fields and environments. RESULTS: Here, we present an approach that develops a deep convolutional neural network (CNN) based on the tiny YOLOv3 architecture for C. sepium and sugar beet detection. We generated 2271 synthetic images, before combining these images with 452 field images to train the developed model. YOLO anchor box sizes were calculated from the training dataset using a k-means clustering approach. The resulting model was tested on 100 field images, showing that the combination of synthetic and original field images to train the developed model could improve the mean average precision (mAP) metric from 0.751 to 0.829 compared to using collected field images alone. We also compared the performance of the developed model with the YOLOv3 and Tiny YOLO models. The developed model achieved a better trade-off between accuracy and speed. Specifically, the average precisions (APs@IoU0.5) of C. sepium and sugar beet were 0.761 and 0.897 respectively with 6.48 ms inference time per image (800 × 1200) on a NVIDIA Titan X GPU environment. CONCLUSION: The developed model has the potential to be deployed on an embedded mobile platform like the Jetson TX for online weed detection and management due to its high-speed inference. It is recommendable to use synthetic images and empirical field images together in training stage to improve the performance of models.

Development of a Stereovision-Based Technique to Measure the Spread Patterns of Granular Fertilizer Spreaders.

Centrifugal fertilizer spreaders are by far the most commonly used granular fertilizer spreader type in Europe. Their spread pattern however is error-prone, potentially leading to an undesired distribution of particles in the field and losses out of the field, which is often caused by poor calibration of the spreader for the specific fertilizer used. Due to the large environmental impact of fertilizer use, it is important to optimize the spreading process and minimize these errors. Spreader calibrations can be performed by using collection trays to determine the (field) spread pattern, but this is very time-consuming and expensive for the farmer and hence not common practice. Therefore, we developed an innovative multi-camera system to predict the spread pattern in a fast and accurate way, independent of the spreader configuration. Using high-speed stereovision, ejection parameters of particles leaving the spreader vanes were determined relative to a coordinate system associated with the spreader. The landing positions and subsequent spread patterns were determined using a ballistic model incorporating the effect of tractor motion and wind. Experiments were conducted with a commercial spreader and showed a high repeatability. The results were transformed to one spatial dimension to enable comparison with transverse spread patterns determined in the field and showed similar results.

Reduction of ammonia emissions from dairy cattle cubicle houses via improved management- or design-based strategies: A modeling approach.

Given the current scarcity of empirical data on ammonia (NH3) emissions from dairy cattle under different management-based mitigation techniques, a modeling approach to assess potential NH3 emission reduction factors is needed. This paper introduces a process-based model that estimates NH3 emission reduction factors for a dairy cattle barn featuring single or multiple management-based NH3 emission mitigation techniques, as compared to another barn, to which no mitigation measure is applied. The model accounts for the following emission mitigation measures: (a) floor scraping, (b) floor type, (c) floor flushing with water and (d) indoor acidification of manure. Model sensitivity analysis indicated that manure acidification was the most efficient NH3 emission reduction technique. A fair agreement was observed between reduction factors from the model and empirical estimates found in the literature. We propose a list of combinations of techniques that achieve the largest reductions. In order of efficiency, they are: (a) floor scraping combined with manure acidification (reduction efficiency 44-49%); (b) solid floor combined with scraping and flushing (reduction efficiency 21-27%); (c) floor scraping combined with flushing and (d) floor scraping alone (reduction efficiency 17-22%). The model is currently being used to advise the Flemish Government (Belgium), on the performance of certain NH3 emission reduction systems for dairy barns in Flanders.

Spray Droplet Characterization from a Single Nozzle by High Speed Image Analysis Using an In-Focus Droplet Criterion.

Accurate spray characterization helps to better understand the pesticide spray application process. The goal of this research was to present the proof of principle of a droplet size and velocity measuring technique for different types of hydraulic spray nozzles using a high speed backlight image acquisition and analysis system. As only part of the drops of an agricultural spray can be in focus at any given moment, an in-focus criterion based on the gray level gradient was proposed to decide whether a given droplet is in focus or not. In a first experiment, differently sized droplets were generated with a piezoelectric generator and studied to establish the relationship between size and in-focus characteristics. In a second experiment, it was demonstrated that droplet sizes and velocities from a real sprayer could be measured reliably in a non-intrusive way using the newly developed image acquisition set-up and image processing. Measured droplet sizes ranged from 24 µm to 543 µm, depending on the nozzle type and size. Droplet velocities ranged from around 0.5 m/s to 12 m/s. The droplet size and velocity results were compared and related well with the results obtained with a Phase Doppler Particle Analyzer (PDPA).

Wind tunnel study of ammonia transfer from a manure pit fitted with a dairy cattle slatted floor.

In dairy cattle systems, most of the feces and urine go to the pit. At the manure pit level, mass transfer of NH3 ([Formula: see text]) has many factors, but practical difficulties hamper a controlled field evaluation. In this study, we propose a methodology for the determination of an alternative, more practical, pit transfer coefficient of NH3 (PTC), and compare it with [Formula: see text] determined from other scientific studies. The aims of this research study were: (1) to develop a wind tunnel set-up which mimics air flow patterns between the slats and above a clean section of a slatted floor section, featuring an aqueous NH3-emitting solution; and (2) to assess how air velocity, turbulence intensity, NH3 concentration ([NH3]) and PTC are influenced by inlet airflow ventilation rate (VR) forced deflection of the air above the slats into the manure pit through varying the deflection angle (DA) of a deflection panel and varying pit headspace height (HH). Main conclusions were: (1) the calculated PTC values presented a good fit to the power function of the air speed near the slats (u) (p < .001) while the average PTC (0.0039 m s(-1)) was comparable to [Formula: see text] values obtained from other studies, by remaining within the range of average values of 0.0015-0.0043 m s(-1); (2) VR and DA significantly impacted [NH3] profiles and PTC (p < .001) and (3) changing slurry pit from 0.10 to 0.90 m HH did not significantly impact [NH3] or PTC (p = .756 and p = .854, respectively).

Development of a High Irradiance LED Configuration for Small Field of View Motion Estimation of Fertilizer Particles.

Better characterization of the fertilizer spreading process, especially the fertilizer pattern distribution on the ground, requires an accurate measurement of individual particle properties and dynamics. Both 2D and 3D high speed imaging techniques have been developed for this purpose. To maximize the accuracy of the predictions, a specific illumination level is required. This paper describes the development of a high irradiance LED system for high speed motion estimation of fertilizer particles. A spectral sensitivity factor was used to select the optimal LED in relation to the used camera from a range of commercially available high power LEDs. A multiple objective genetic algorithm was used to find the optimal configuration of LEDs resulting in the most homogeneous irradiance in the target area. Simulations were carried out for different lenses and number of LEDs. The chosen configuration resulted in an average irradiance level of 452 W/m² with coefficient of variation less than 2%. The algorithm proved superior and more flexible to other approaches reported in the literature and can be used for various other applications.

High speed stereovision setup for position and motion estimation of fertilizer particles leaving a centrifugal spreader.

A 3D imaging technique using a high speed binocular stereovision system was developed in combination with corresponding image processing algorithms for accurate determination of the parameters of particles leaving the spinning disks of centrifugal fertilizer spreaders. Validation of the stereo-matching algorithm using a virtual 3D stereovision simulator indicated an error of less than 2 pixels for 90% of the particles. The setup was validated using the cylindrical spread pattern of an experimental spreader. A 2D correlation coefficient of 90% and a Relative Error of 27% was found between the experimental results and the (simulated) spread pattern obtained with the developed setup. In combination with a ballistic flight model, the developed image acquisition and processing algorithms can enable fast determination and evaluation of the spread pattern which can be used as a tool for spreader design and precise machine calibration.

The HMI™ module: a new tool to study the Host-Microbiota Interaction in the human gastrointestinal tract in vitro.

BACKGROUND: Recent scientific developments have shed more light on the importance of the host-microbe interaction, particularly in the gut. However, the mechanistic study of the host-microbe interplay is complicated by the intrinsic limitations in reaching the different areas of the gastrointestinal tract (GIT) in vivo. In this paper, we present the technical validation of a new device--the Host-Microbiota Interaction (HMI) module--and the evidence that it can be used in combination with a gut dynamic simulator to evaluate the effect of a specific treatment at the level of the luminal microbial community and of the host surface colonization and signaling. RESULTS: The HMI module recreates conditions that are physiologically relevant for the GIT: i) a mucosal area to which bacteria can adhere under relevant shear stress (3 dynes cm(-2)); ii) the bilateral transport of low molecular weight metabolites (4 to 150 kDa) with permeation coefficients ranging from 2.4 × 10(-6) to 7.1 × 10(-9) cm sec(-1); and iii) microaerophilic conditions at the bottom of the growing biofilm (PmO2 = 2.5 × 10(-4) cm sec(-1)). In a long-term study, the host's cells in the HMI module were still viable after a 48-hour exposure to a complex microbial community. The dominant mucus-associated microbiota differed from the luminal one and its composition was influenced by the treatment with a dried product derived from yeast fermentation. The latter--with known anti-inflammatory properties--induced a decrease of pro-inflammatory IL-8 production between 24 and 48 h. CONCLUSIONS: The study of the in vivo functionality of adhering bacterial communities in the human GIT and of the localized effect on the host is frequently hindered by the complexity of reaching particular areas of the GIT. The HMI module offers the possibility of co-culturing a gut representative microbial community with enterocyte-like cells up to 48 h and may therefore contribute to the mechanistic understanding of host-microbiome interactions.

Effect of spray angle and spray volume on deposition of a medium droplet spray with air support in ivy pot plants.

BACKGROUND: Spray boom systems, an alternative to the predominantly-used spray guns, have the potential to considerably improve crop protection management in glasshouses. Based on earlier experiments, the further optimization of the deposits of a medium spray quality extended range flat fan nozzle type using easy adjustable spray boom settings was examined. Using mineral chelate tracers and water sensitive papers, the spray results were monitored at three plant levels, on the upper side and the underside of the leaves, and on some off-target collectors. In addition, the deposition datasets of all tree experiments were compared. RESULTS: The data showed that the most efficient spray distribution with the medium spray quality flat fan nozzles was found with a 30° forward angled spray combined with air support and an application rate of 1000 L ha(-1) . This technique resulted in a more uniform deposition in the dense canopy and increased spray deposition on the lower side of the leaves compared with the a standard spray boom application. Applying 1000 L ha(-1) in two subsequent runs instead of one did not seem to show any added value. CONCLUSION: Spray deposition can be improved hugely simply by changing some spray boom settings like nozzle type, angling the spray, using air support and adjusting the spray volume to the crop.

Effect of spray application technique on spray deposition in greenhouse strawberries and tomatoes.

BACKGROUND: Increasingly, Flemish greenhouse growers are using spray booms instead of spray guns to apply plant protection products. Although the advantages of spray booms are well known, growers still have many questions concerning nozzle choice and settings. Spray deposition using a vertical spray boom in tomatoes and strawberries was compared with reference spray equipment. Five different settings of nozzle type, size and pressure were tested with the spray boom. RESULTS: In general, the standard vertical spray boom performed better than the reference spray equipment in strawberries (spray gun) and in tomatoes (air-assisted sprayer). Nozzle type and settings significantly affected spray deposition and crop penetration. Highest overall deposits in strawberries were achieved using air-inclusion or extended-range nozzles. In tomatoes, the extended-range nozzles and the twin air-inclusion nozzles performed best. Using smaller-size extended-range nozzles above the recommended pressure range resulted in lower deposits, especially inside the crop canopy. CONCLUSIONS: The use of a vertical spray boom is a promising technique for applying plant protection products in a safe and efficient way in tomatoes and strawberries, and nozzle choice and setting should be carefully considered.

Long-term trends in the prevalence of cancer and other major diseases among flatfish in the southeastern North Sea as indicators of changing ecosystem health.

This paper analyses and discusses spatial and temporal patterns in the prevalence of major skin diseases (lymphocystis, epidermal hyperplasia/papilloma, ulcers), intestinal parasite Glugea sp., and liver cancer in dab (Limanda limanda) and flounder (Platichthys flesus) in the Dutch section of the North Sea since the mid-1980s. We have attempted to relate disease prevalence trends in both species to chemical contaminant exposure and other relevant environmental factors including fish condition factor, population density, fishing activity, and water temperature. We observed a long-term decline in chemical-related liver cancer in the populations of both species since the early 1990s. Lymphocystis and skin ulcer (flounder only) have also displayed a significant decrease since then. We conclude that the widespread decline in the prevalence of several skin diseases and liver cancer in dab and flounder in Dutch waters in the past two decades is most likely due to the improved water quality and health conditions in this region.

Technical and economic feasibility of gradual concentric chambers reactor for sewage treatment in developing countries

A major challenge in developing countries concerning domestic wastewaters is to decrease their treatment costs. In the present study, a new cost-effective reactor called gradual concentric chambers (GCC) was designed and evaluated at lab-scale. The effluent quality of the GCC reactor was compared with that of an upflow anaerobic sludge bed (UASB) reactor. Both reactors showed organic matter removal efficiencies of 90 percent; however, the elimination of nitrogen was higher in the GCC reactor. The amount of biogas recovered in the GCC and the UASB systems was 50 percent and 75 percent of the theoretical amount expected, respectively, and both reactors showed a slightly higher methane production when the feed was supplemented with an additive based on vitamins and minerals. Overall, the economical analysis, the simplicity of design and the performance results revealed that the GCC technology can be of particular interest for sewage treatment in developing countries.

Accelerated solid-phase dynamic extraction of toluene from air.

Solid-phase dynamic extraction (SPDE) belongs to the most innovative sample preparation and enrichment techniques. However, there is still a lack of knowledge on the fundamentals of SPDE and its applicability in the field of environmental monitoring. A homemade sampling device is constructed to make a detailed study of SPDE kinetics for toluene extraction. It proved that at least 50 aspirating and dispensing cycles were necessary to obtain toluene equilibration between gas and coating phase. A mechanistic model is proposed to explain that in every dispensing step during SPDE, significant losses of retained analytes (up to 48%) occur due to desorption processes. A new accelerated solid-phase dynamic extraction procedure (ASPDE) has been developed that avoids dispensing stages during extraction. The resulting extraction time proves to be 1.7 min, being a reduction by a factor of 37 compared with the SPDE extraction time. ASPDE proved to have high potential in ambient/indoor air monitoring. The limit of detection for toluene was determined to be 56 ppb(v), i.e. a factor of respectively, 6 and 35 lower than obtained with SPME and conventional headspace sampling with gas syringe.

Computational fluid dynamics analysis for process impact assessment during thermal pasteurization of intact eggs.

Transient temperature and albumen velocity profiles during thermal pasteurization of intact eggs were studied using a commercial computational fluid dynamics (CFD) package. Simulated temperature profiles were in close agreement with experimental data for eggs of different sizes. Convective heat transfer only occurred in the egg white fraction, and conductive heat transfer only occurred in the yolk. For process assessment, a generally accepted kinetic inactivation model for Salmonella Enteritidis was incorporated into the CFD analysis. Minimum process times and temperatures needed to provide equivalent pasteurization at 5-log reductions of the target microorganism were obtained on a theoretical basis. The combination of CFD analysis and inactivation kinetics can be very useful for assessing pasteurization of intact eggs and can enable processors to gain a better understanding of these processes and to establish process conditions for consumer-safe eggs.