AgriEngineering, Volume 8, Issue 3 (March 2026) – 43 articles

Assessing crop water status during the reproductive development of winter wheat is challenging because soil–plant–atmosphere interactions are strongly influenced by soil physical conditions, and measured soil water content (SWC) does not necessarily reflect plant-accessible water. This study applied an integrated, process-based multisensor approach to evaluate functional crop water status and its relationship to grain yield, combining hyperspectral canopy reflectance, atmospheric observations, in situ SWC, and pedological characterization. Five winter wheat cultivars were monitored at two contrasting pedoclimatic sites in continental Croatia during the 2022/2023 growing season. Hyperspectral canopy reflectance (350–2500 nm) was measured at reproductive stages (BBCH 61–83), and seventeen vegetation indices describing canopy water status, structure, pigments, and senescence were derived. Principal component analysis (PCA) identified location as the dominant source of spectral variability, while cultivar effects were secondary. Although atmospheric conditions were broadly comparable, the sites differed markedly in soil physical properties, resulting in contrasting soil water–air regimes. Despite consistently higher volumetric SWC at one site, hyperspectral indicators revealed lower canopy water status, reduced canopy structure, earlier senescence, and lower grain yield across all cultivars. Water-sensitive indices exploiting near-infrared (700–1300 nm) and shortwave infrared (1300–2400 nm) bands (NDWI, NDMI, NMDI, MSI) consistently indicated greater physiological stress. Conversely, the site with lower SWC but more favorable soil physical conditions exhibited higher values of water- and structure-related indices and achieved higher grain yield, with a mean increase of 669 kg ha⁻¹. The results demonstrate that hyperspectral canopy reflectance captures yield-relevant water stress that cannot be inferred from soil moisture alone, highlighting the importance of multisensor integration for interpreting soil–plant–atmosphere interactions under heterogeneous soil conditions. Full article

Data-driven approaches are increasingly required to optimize biofertilization strategies in forage systems. Machine learning (ML) provides an efficient tool for identifying functional drivers in complex plant–soil–microbe systems, offering important perspectives for precision data-driven agriculture. However, despite its potential, ML remains data-driven in studies involving diazotrophic inoculation using biochar as a pelletizing material, particularly in forage grasses. This study applied ML to predict the key drivers controlling Brachiaria brizantha performance and soil quality under biochar-pelletized diazotrophic bacteria (DB). Five isolates were inoculated with or without biochar, and plant traits and soil attributes, including pH, potassium, phosphorus, sodium, and urease activity were evaluated. These data were integrated into multivariate analyses and ML algorithms, including Linear Discriminant Analysis, Random Forest, and Support Vector Machine, to identify the functional drivers that best discriminate treatment performance and uncover mechanistic functional drivers. All isolates increased soil potassium content, with the highest values in the biochar amended treatments, and a 39% increase. Soil pH and urease activity were significantly modulated by isolate identity, while biomass allocation patterns differed among treatments. Overall, the results highlight that biochar pelletization can enhance the effectiveness of DB inoculants. ML revealed that dry foliar biomass, soil pH, and fresh root weight were the most predictive variables, highlighting consistent signatures explaining plant–soil responses to biochar-pelletized DB. These findings demonstrate that interpretable ML can disentangle complex plant–soil–microbe interactions, support precision biofertilization design, and serve as an efficient decision-support tool for sustainable pasture management. Beyond the present system, this study establishes a transferable and scalable analytical framework for precision biofertilization strategies in forage systems and other biochar-mediated agroecosystems, advancing predictive and data-driven approaches in sustainable agricultural engineering. Full article

Feed handling activities in cattle feedlots—such as feed leveling, reconditioning, and the removal of feed refusals along the trough—remain largely manual in many developing countries, resulting in high labor demand and inconsistent feed availability. This study aimed to design, develop, and evaluate an integrated machine capable of leveling feed, conditioning refused feed, and collecting feed refusals. The machine was developed using trough-geometry data (550–650 mm) and feed-residue properties (particle size, bulk density, terminal velocity), integrating a shovel–dual-brush unit with pneumatic suction. A prototype was subsequently fabricated and tested under practical feedlot conditions using various trough widths and operating speeds. Performance evaluation included feed-pile geometry, distribution uniformity, suction efficiency, suction capacity, and fuel consumption. The leveling mechanism significantly improved feed distribution uniformity, reducing the coefficient of variation for feed-pile height and feed mass by up to 67% and 73%, respectively. During conditioning, the machine increased feed-pile height by 8–10 cm and reduced pile width by 6–13 cm. The suction system maintained high efficiency (94–97%) with an average capacity of 14.1 ± 0.8 kg·min⁻¹. Fuel consumption ranged from 0.54 L·h⁻¹ during leveling to 1.30 L·h⁻¹ during suction. Overall, the machine offers a practical solution for improving feed shaping, uniformity, and residue removal. Full article

Traditional mechanized chestnut harvesting is too costly for small producers, non-selective, and prone to damaging nuts. Accurate, reliable detection of chestnuts on the orchard floor is crucial for developing low-cost, vision-guided automated harvesting technology. However, developing a reliable chestnut detection system faces challenges in complex environments with shading, varying natural light conditions, and interference from weeds, fallen leaves, stones, and other foreign on-ground objects, which have remained unaddressed. This study collected 319 images of chestnuts on the orchard floor, containing 6524 annotated chestnuts. A comprehensive set of 29 state-of-the-art real-time object detectors, including 14 in the YOLO (v11–v13) and 15 in the RT-DETR (v1–v4) families at various model scales, was systematically evaluated through replicated modeling experiments for chestnut detection. Experimental results show that the YOLOv12m model achieved the best mAP@0.5 of 95.1% among all the evaluated models, while RT-DETRv2-R101 was the most accurate variant among the RT-DETR models, with mAP@0.5 of 91.1%. In terms of mAP@[0.5:0.95], the YOLOv11x model achieved the best accuracy of 80.1%. All models demonstrated significant potential for real-time chestnut detection, and YOLO models outperformed RT-DETR models in terms of both detection accuracy and inference, making them better suited for on-board deployment. This work lays a foundation for developing AI-based, vision-guided intelligent chestnut harvest systems. Full article

The proposed study investigates the key factors influencing UAV adoption and proposes an integrated educational–operational framework to enhance implementation in agricultural practice. A case study in Sibiu County, Romania, combined survey-based empirical analysis (n = 80), strategic environmental assessment and the deployment of a demonstration aerial crop monitoring system for educational purposes (ACMS-E). We integrated the Technology Acceptance Model (TAM) and Theory of Planned Behavior (TPB) to examine adoption intentions, revealing perceived usefulness (β = 0.355, p = 0.021) and positive attitudes (β = 0.382, p = 0.005) as the strongest predictors, explaining 44.1% of variance. Based on these findings, a modular training curriculum was designed, combining theoretical instruction, flight operation exercises, remote sensing techniques, data analytics and farm-management integration. ACMS-E provides hands-on training and promotes capacity-building, bridging the gap between technological availability and real-world adoption. By linking technological capabilities with structured training, ACMS-E bridges the gap between UAV availability and effective implementation, offering a scalable model for precision agriculture. This framework provides a pathway to accelerate UAV adoption, optimize field-level monitoring and support evidence-based, resource-efficient farm management in emerging and developed agricultural contexts. Full article

Driven by the global rise in animal protein demand, poultry farming has evolved into a highly intensive and technically complex sector. According to the FAO, animal protein production increased by about 16% in the past decade, with poultry alone expanding by 27% and becoming the leading source of animal protein. This intensification requires rapid, complex decisions across multiple aspects of production under uncertainty and strict time constraints. This study presents the development and evaluation of a conversational decision support system (DSS) designed to support decision-making to assist poultry producers, particularly broiler producers, in addressing technical queries across five key domains: environmental control, nutrition, health, husbandry, and animal welfare. As a proof-of-concept study, the reference context is intensive broiler production, covering common floor-rearing housing settings, including environmentally controlled and mechanically ventilated houses. The system combines a large language model (LLM) with retrieval-based generation (RAG) to ground responses in a curated corpus of scientific and technical literature. Additionally, it adds a reasoning component using Paraconsistent Annotated Evidential Logic Eτ, a non-classical logic designed to handle contradictory or incomplete information. Methodologically, Logic Eτ is used as a workflow-level control mechanism to gate clarification, domain routing, and answer adequacy signaling, rather than serving only as a post hoc label on generated outputs. Evaluation was conducted by comparing system responses with expert reference answers using semantic similarity (cosine similarity with SBERT embeddings). The results indicate that the system successfully retrieves and composes relevant content, while the paraconsistent inference layer makes results easier to interpret and more reliable in the presence of conflicting or insufficient evidence. These findings suggest that the proposed architecture provides a viable foundation for explainable and reliable decision support in modern poultry production, achieving consistent reasoning under contradictory or incomplete information where conventional RAG chatbots may produce unstable guidance. Full article

It is necessary to understand the complicated interplay between diseases and medicinal plants to find new curing agents that may be used in natural sources. Nevertheless, the state of interaction between diseases and plants today is not fully developed yet, and the potentially productive plant-based treatment can hardly be identified rationally. In order to elaborate on this challenge, we will offer a heterogeneous network approach to the prediction of novel disease–plant associations by using the Random Walk with Restart (RWR) algorithm. The framework combines three significant relational networks, including (i) a disease–plant association network, which has been built using curated literature and biological databases, (ii) a disease–disease similarity net, which is constructed using shared symptoms and therapeutic profiles, and (iii) a plant–plant similarity net using phytochemical and functional similarities. These elements are integrated into a homogeneous graph that is heterogeneous in nature, and thus, information flows through related nodes. The model begins by finding RWR between known disease or plant nodes and develops the network by exploring the graph further to make estimates of the probability of association between disease and plant networks that were not previously connected. Experimental tests show that the proposed model has an excellent predictive ability, ROC-AUC of 0.9987, PR-AUC equal to 0.915, and Precision = 10 of 1.0, significantly better than the results of the base models, including Random- and Degree-based models. The bootstrap analysis supported the strength of the model as the mean ROC-AUC was 0.9987 with a standard deviation of 0.00051. The suggested structure offers an effective computational methodology to systematically explore disease–plant interactions to aid in finding novel herbal drugs to treat diseases and speed up the drug discovery process by means of inference based on networks. Full article

Driven by agricultural labor shortages and rising quality requirements, ginger harvesting increasingly demands high-throughput, low-damage operations and a reliable supply chain. This review summarizes harvesting modes and harvester types used in ginger production, with emphasis on critical process modules: digging and lifting, soil disintegration and cleaning, vine cutting and anti-tangling, gentle conveying, and collection. We compare major technical routes in terms of field capacity, control of soil and foreign materials, damage mitigation, and reliability under continuous operation, and identify the conditions under which each route performs best. Drawing on advances in harvesting systems for other root and bulb crops, we outline transferable approaches for intelligent sensing, precision control, and system-level integration. We then propose an online monitoring and closed-loop regulation framework for strongly coupled conditions, such as heavy clay soils, plastic-mulch residues, and vine interference. Key bottlenecks include limited cross-regional adaptability, persistent trade-offs between low damage and high throughput, cost constraints on intelligent functions, and the lack of shared datasets and standardized evaluation protocols. Future progress should be anchored in integrated equipment sets and supporting operating specifications, guided by multi-source sensing-based quality indicators and interpretable control strategy libraries, to reduce harvest losses, stabilize marketable quality, improve operational efficiency, and enable scalable adoption. Full article

Sheep farming length of stay in the feedlot directly influences system profitability, mainly due to the high cost of feed. Thus, the use of predictive models based on body measurements is an important tool to define the optimal slaughter point and the ideal feedlot period. Thus, the aim was to evaluate predictive models of meat cuts and tissue carcasses concerning weight at slaughter (WS), loin eye area (LEA), and subcutaneous fat thickness (SFT) obtained by ultrasound of the lumbar region of lambs. The WS and ultrasound measurements were obtained from a pre-slaughter collection of 45 lambs, divided into five groups, each weighing 15, 20, 25, 30, or 35 kg, with nine replications per group. Three regression models were evaluated: WS, LEA, and SFT (independent variables) and the cuts yield or tissue composition (dependent variable). Increasing WS resulted in greater carcass weight and commercial cuts. Above 15 kg body weight, bone weight showed little or no increase (allometric coefficient = 0.06), whereas muscle and fat tissues increased steadily, with allometric coefficients of 0.25 and 0.12, respectively. The commercial cuts showed a high and significant correlation with WS and LEA. The muscle and bone proportion of the leg had a significant (p < 0.10) correlation with SFT. For the weight of commercial cuts estimates, the inclusion of LEA and/or SFT with WS did not improve the coefficient of determination but made the predictions equivalent to the measured values. There were high determination coefficients when WS was only used to predict muscle, fat, and bone weight, but it was not efficient in predicting the muscle/fat and muscle/bone ratios and the percentage of tissues. The WS was the variable that best explained the weight and tissue content. The inclusion of LEA and/or SFT made little improvement to the predictive models. Full article

The current research undertook a comprehensive examination of global research related to the use of measurement, reporting, and verification (MRV) techniques for quantifying and tracking greenhouse gas (GHG) emissions from agriculture and livestock farming. Data were collected using a bibliometric analysis of 5340 studies published in the period (1990–2025) and a systematic literature review of 100 studies published in the period (2020–2025). The insights from the findings showed that four MRV techniques were broadly adopted across different regions: (1) inventory techniques (IPCC Tiers, national systems), (2) accounting at the project/product level (LCA, carbon footprint protocols), (3) MRV based on measurement and models (chambers, remote sensing, farm models, AI/ML), and (4) frameworks for governance and standardization (UNFCCC, Paris ETF, PAS 2050, etc.). The findings further revealed the impact of the MRV techniques on agriculture and livestock farming, showing that they facilitated the uptake of low-carbon practices. In agriculture, the MRV techniques showed that lower emissions emerged from mixed cropping, while in livestock farming, the emissions varied based on the feeding stage and type of diet used. However, various challenges arose in the adoption of MRV techniques where there was limited data related to GHG emissions, thereby reducing generalizability. In future work, there is a need for scholars to consider integrating the different MRV techniques to develop an understanding of the problem area. Full article

Crop monitoring over large land extensions represents a central challenge in precision agriculture, especially in polyculture contexts where species with different nutritional needs are combined. This study presents a methodology to manage and analyze large volumes of multispectral images captured by unmanned aerial vehicles (UAVs) in order to identify and monitor crops at the plant level. The images are efficiently stored and retrieved using a Hilbert Curve, which reduces the complexity of the search process from $O\left(n^2\right)$ to $O(log(n\left)\right)$ where n represents the number of indexed data points). The system connects to a distributed Structured Query Language (SQL) database, allowing for fast image retrieval based on GPS coordinates and other metadata. Additionally, the Normalized Difference Vegetation Index (NDVI) is calculated using reflectance data from the red and near-infrared channels, adjusted by semantic segmentation masks generated with a U-Net model, which allows for species-specific evaluations. The methodology was evaluated on a 20,000 m² polyculture farm with coffee, avocado, and plantain crops, using a dataset of 270 aerial images partitioned into 70% for training and 30% for validation. The results show improvements in retrieval speed and precision with the Hilbert Space-Filling Curve (HSFC) approach, and an accuracy of 82.3% and an the Mean Intersection over Union (MIoU) of 68.4% in species detection with the U-Net model. Overall, this integrated framework demonstrates a scalable potential for precision agriculture in complex polyculture systems, facilitating efficient data management and targeted crop interventions. Full article

The growing demand for sustainable soil management strategies has intensified interest in hydrochar (HC), a waste-derived amendment produced via hydrothermal carbonization (HTC). This review synthesizes recent advances in HC production, characterization, and agri-environmental applications within a waste-to-resource framework. It covers studies conducted mainly over the last decade, encompassing a wide range of feedstocks, including agricultural residues, sewage sludge, animal manures, and food waste. HTC is typically performed at 130–280 °C under autogenous pressure (2–15 MPa), generating HCs with low intrinsic surface area (<50 m²g⁻¹) and oxygen-containing functional groups that govern nutrient dynamics and soil interactions. Reported application rates vary broadly between 10 and 60 t ha⁻¹, with most experiments conducted under greenhouse conditions. Positive effects on soil pH, cation exchange capacity, water retention, and phosphorus availability are frequently observed. However, plant responses vary according to the type of stimulation promoted by HC, as well as its processing conditions, application rates, and the soil characteristics in which it is applied. Advanced molecular-level analyses (e.g., FT-ICR-MS, GC-MS, and ¹³C-NMR) have provided mechanistic insights into carbon stability, nutrient release, and interaction with soil organic matter. Reusing HTC process water offers an additional pathway for nutrient recovery, although concerns about phytotoxic compounds remain. Despite promising short-term results, long-term field evaluations and standardized assessment protocols are still limited. This review integrates structural, functional and agri-environmental perspectives to identify critical knowledge gaps and guide the optimized and context specific use of hydrochar in sustainable agricultural systems. At the same time, it emphasizes its role in advancing carbon sequestration and in operationalizing resource-circular strategies, thereby underscoring its broader practical and strategic relevance. Full article

To optimize the structural parameters of a peanut digging shovel and enhance its operational performance, the forces exerted on the digging shovel were examined through a graphical mechanics approach. This analysis identified the primary structural and operational parameters of the shovel’s design. A numerical simulation model for the working resistance of the shovel was established adopting EDEM (2018) discrete element analysis software and subsequently validated through comparative analysis with field experiment results. Employing the Box–Behnken response surface method, quadratic regression models were constructed with digging resistance and soil non-breakage ratio as the response variables, while forward speed, soil entry angle, and blade tilt angle were taken as the influencing factors. Optimization analysis of these parameters was conducted. The optimization results indicate that with a forward speed of 0.8 m/s, a soil entry angle of 20°, and a blade tilt angle of 40°, the working resistance of the shovel is 1667 N, and the soil non-breakage ratio is 20.56%. The error between the field test results and the predictions from the optimized model was less than 2%, illustrating the feasibility of the model and the optimization outcomes. This study offers a technical reference for future simulation-based optimization of peanut digging shovels. Full article

Mulch-film covering technology has been widely adopted in cotton production in arid regions; however, the associated problem of residual-film pollution has become increasingly prominent, creating an urgent demand for efficient and accurate monitoring approaches. Owing to the small target scale, irregular morphology, blurred boundaries, and complex soil backgrounds of residual-film fragments, residual-film detection based on close-range UAV imagery remains a challenging task. To address these issues, this study proposes an improved algorithm, termed YOLO-EGASF, for residual-film detection in arid-region cotton fields, built upon the lightweight YOLOv11n framework. To enhance the detection of small targets with weak boundary characteristics, the baseline model is improved from three aspects. First, a boundary-enhanced multi-branch small-target extraction module (EMSE) is designed to reinforce shallow-layer details and gradient information through multi-scale convolution and explicit edge enhancement. Second, a GLoCA attention module that integrates global coordinate information with local geometric features is constructed to improve the discriminative capability of the model for residual-film targets under complex background conditions. Third, an ASF-layer multi-scale feature fusion structure is introduced, together with an additional small-target detection layer, to strengthen the participation of high-resolution features in cross-scale fusion and prediction. Experimental results on a self-constructed UAV-based residual-film dataset from cotton fields demonstrate that YOLO-EGASF outperforms several mainstream detection models in terms of Precision, Recall, mAP@0.5, and mAP@0.5:0.95, achieving mAP@0.5 and mAP@0.5:0.95 values of 71.9% and 26.8%, respectively. These results indicate a significant improvement in detection accuracy and robustness, confirming that the proposed method can effectively meet the practical requirements of fine-grained residual-film monitoring in arid-region cotton fields. Full article

The mechanization of the agricultural sector exposes operators to vibrations generated by tractors, terrain inclination, and attached implements. Prolonged exposure to such vibrations can lead to health problems, including visual disturbances, fatigue, spinal injuries, and low back pain. In this context, the present study aimed to assess the severity of mechanical vibrations in an agricultural tractor with four-wheel drive, both as a standalone unit and as part of a mechanized assembly comprising the same tractor coupled to a fertilizer seeder during sowing operations. Vibrations were monitored at four data collection points: the front and rear axles, the cab floor, and the operator’s seat. Root mean square (RMS) acceleration values were compared with the limits established by ISO 2631-1, and the comfort levels at the operator’s seat were classified as “uncomfortable” and “very uncomfortable.” Vibration transmissibility between the rear axle and the cab floor (T2) was found to exceed 1, indicating amplification of vibrations. Overall, the operator’s seat attenuated the vibration severity transmitted to the operator. Forward speed significantly influenced vibration severity, with higher speeds associated with increased RMS accelerations. Slope also affected vibration levels, with slope D2 (the sloped area) presenting higher mean RMS acceleration values. Notably, the tractor operating with the seeder fertilizer exhibited attenuated vibration levels compared to the tractor alone. Full article

The increasing demand for agricultural products, intensified by natural resource degradation and the lack of human labor in the agri-food sector, favors the adoption of advanced automated technologies in the entire farm-to-fork chain. Despite skepticism, 3D (three-dimensional) printing is amongst the methods that have drawn increasing attention and encourage expectations for tackling the aforementioned challenges. In this context, the current work has a multiperspective character. Firstly, it sheds light on the recent progress in the 3D printing fabrication area and focuses on laboratory-implemented parts improving the efficiency of typical agricultural processes. These cost-effective solutions vary from covers for damaged electric water pumps and joints for greenhouse structures to adjustable ventilation grilles, automatic irrigation valves and specialized fruit-harvesting grippers. Secondly, it reports on lessons learned, highlighting potential strengths/weaknesses during the fabrication process, assisted by complementary feedback collected via questionnaires from agricultural engineering students, their professors, and farmers. Experiences gained justify the optimism about the capacity of 3D printing to foster agriculture, but there are still concerns about the easiness of the 3D printing process and the ability of the 3D-printed parts to withstand harsh agricultural field conditions. Finally, it indicates future directions for the incorporation of 3D printing in agriculture toward increased sustainability pathways. Full article

Goats are important livestock species in most rural households and were amongst the first species to be domesticated. Despite this, their production is based on extensive systems, exposing them to numerous challenges affecting their productivity. This review examines the applications of precision livestock farming (PLF) and AI-driven technologies in goat management, focusing on their impacts on productivity, welfare, genetic potential, health monitoring, feeding efficiency and sustainability outcomes and identifying challenges for their adoption in smallholder and extensive systems. Unlike previous reviews that focus mainly on cattle raised under intensive systems, this review synthesizes their use in goat production and highlights technological, socio-economic and infrastructural constraints. A conventional literature review approach is used, with studies retrieved from major databases using relevant keywords. The selected studies are evaluated to assess technological applications, benefits and adoption challenges, followed by a SWOT analysis. Engineering aspects of precision livestock farming—including sensors, data connectivity, system integration, automation and scalability—are also discussed. Ideally, these technologies operate as integrated decision-support systems that jointly improve productivity, animal welfare and sustainability, rather than performing isolated tasks. However, many PLF solutions remain at low technology-readiness levels and are constrained by infrastructure gaps, sensor reliability and compatibility issues, which collectively limit adoption in smallholder systems. Future research should focus on the development of cost-effective, reliable PLF systems for smallholder producers, while policy and capacity-building initiatives are needed to enhance infrastructure, training and technology adoption for scalable implementation. Full article

Wind is a primary driver of nonuniform water application in traveling gun sprinklers, yet design guidance often treats wind only as speed. This study quantifies how diurnal wind regimes (day vs. night) and wind incidence relative to the towpath (φ) affect application-rate patterns and the Christiansen uniformity coefficient (UC) as a function of towpath spacing expressed as a fraction of wetted diameter (WD). Class-specific sprinkler patterns were generated with the Simulation Model for Sprinkler Irrigation (SIA) and combined with local daytime and nighttime wind-frequency data to build composite application-rate fields; these drove traveler simulations that computed cross-track depth, lateral overlap across spacings, and UC for representative wind speeds (0–6 m s⁻¹) and φ (0°, 45°, 90°). Nighttime operation yielded higher UC, with a day–night crossover near ~50% WD and an average UC gain of ~9.5 percentage points; typical gains were +6 to +9 points between 55% and 90% WD. Wind incidence was as influential as speed: at 65.6% WD, increasing wind from 0 to 6 m s⁻¹ reduced UC from 84.4% to 28.6% for φ = 0°, to 52.0% for 45°, and to 76.1% for 90°. Findings support nighttime scheduling, towpaths avoiding wind-parallel operation, and tighter spacings under windy conditions. Full article

The automatic detection of foliar nutritional deficiencies through computer vision represents a promising alternative within precision agriculture practices, reducing dependence on laboratory analyses and the subjectivity associated with visual inspection. This systematic review maps and compares the application of machine learning (ML) and deep learning (DL) techniques to nutritional diagnosis across different crops, highlighting methodological trends, barriers to model adoption under field conditions, and existing research gaps. Following the PRISMA guidelines (PRISMA-P and PRISMA-2020), searches were conducted in the Scopus, IEEE Xplore, and Web of Science databases, using a defined time frame and explicit inclusion and exclusion criteria, resulting in 200 articles included (2012–2026; last search on 2 February 2026). The results indicate a predominance of DL-based approaches and RGB imagery, with applications concentrated in crops such as rice and in macronutrients, mainly nitrogen (N), phosphorus (P), and potassium (K), and report a marked increase in publications from 2020 onward. Although many studies report high performance, the evidence is largely derived from controlled environments and proprietary datasets, which limit model comparability, reproducibility, and generalization to real-world scenarios. Accordingly, the main research gaps include limited validation under field conditions, identified as the primary practical barrier; the underrepresentation of micronutrients and multiple-deficiency diagnosis; and the need for lightweight architectures suitable for deployment in mobile and edge-computing applications. It is concluded that ML and DL techniques offer promising alternatives for automated nutritional diagnosis; however, advances in data standardization, open-access datasets, and validation under real field conditions are essential for consolidating these technologies in practical applications. Full article

Machine learning has enhanced rice grain classification by enabling accurate, automated, and objective morphological analysis, supporting quality control and varietal selection. This study compared the performance of several algorithms in identifying three Ecuadorian rice varieties (INIAP-11, INIAP-12, and INIAP-20) using a balanced dataset of morphological features. Five models were trained with cross-validation and evaluated using multi-class metrics. Significant differences among varieties particularly in area, length, and eccentricity confirmed their discriminative potential. Initially, models were trained using all morphological variables. However, to optimize training time and computational cost, the study also evaluated model performance after applying dimensionality reduction through Principal Component Analysis (PCA). This approach enabled assessing whether reduced feature spaces could maintain competitive predictive performance while improving efficiency. Overall, all algorithms performed well, but only the Artificial Neural Network (ANN) and Support Vector Classifier (SVC) demonstrated strong generalization without overfitting. In contrast, Random Forest achieved perfect accuracy in training but decreased performance in testing. In conclusion, ANN and SVC emerged as the most robust alternatives for rice grain morphological classification, while the PCA results highlight the value of dimensionality reduction as a strategy to enhance computational scalability without substantially compromising accuracy. The objective of the present study is to train, evaluate, and compare different machine learning algorithms for the classification of three types of rice grains, in order to determine the best model for this task based on seven morphological characteristics of the grains applying machine learning algorithms with and without dimensional reduction. Full article

The biospeckle laser (BSL) technique is recognized as a sensitive method for detecting biological activity and has been successfully applied for seed vigor testing. Given these achievements, whether the integration of BSL into automated systems can provide complementary information on the seed imbibition process remains limited. Addressing this gap represents a significant challenge with strong potential for technological innovation. This study presents an automated laser-optical system designed to monitor the imbibition process of multiple seeds over time using a mechanized carousel. The developed apparatus integrates all necessary components for the illumination and image acquisition of eight seeds across programmable time intervals, controlled by an industrial-grade programmable controller. Validation using maize seeds (Zea mays L.) over a 36-h period confirmed the system’s reliability. BSL indices enabled the characterization of internal biological activity throughout imbibition, revealing dynamic processes that remained undetected in previous discrete-time analyses. These results highlight the potential of the proposed system for more comprehensive and continuous seed monitoring. The successful automated laser-optical system with relative humidity control opens great potential in seeds research and daily industrial analysis. The test of the proposed system in other seeds is the next challenge, regarding the thick and colored coats. The design of larger carousels is a possible step forward, which will demand studies of the limits linked to the illumination and image acquisition time performed in each seed. Full article

This article presents a simple and understandable approach to the automatic assessment of the severity of late blight on tomato leaves. We collect our own dataset of 5245 RGB images of healthy and diseased tomato leaves and determine five ordinal classes: healthy (0%) and four infection levels (0.1–10%, 11–25%, 26–50%, and ≥51% of the affected area). Each image is segmented using the global definition of the Otsu threshold, followed by morphological purification, after which seven textural and geometric characteristics are extracted from the contours of the lesion: contrast, number of contours, average and standard deviation of the contour area, average and standard deviation of the contour perimeter, and average area-to-perimeter ratio. All characteristics are normalized and used as input data for the XGBoost classifier. The dataset is randomly split into 80% training and 20% test images, resulting in an independent test set of 1049 images. In this test set, the proposed model provides an overall accuracy of 0.93 and an F1 macro score of 0.93 points, while for each F1 class, it varies from 0.90 to 0.97. The confusion matrix shows a stable difference between neighboring severity levels, while the analysis of the importance of the features confirms the relevance of contour descriptors for characterizing the size and shape of the lesion. This method only runs on a central processor, requires a small amount of memory, and outputs interpretable output data, making it suitable for use in greenhouses and farms with limited computing resources. We also discuss the limitations associated with the boundaries between neighboring classes and the potential shift in the subject area, and we outline directions for expanding the approach to multi-sheet scenes and explicit ordinal loss functions. Full article

The mechanical vibrations that occur in agricultural machinery, arising from terrain irregularities or the moving parts of the machine, can harm the operators when they are subjected to work for many hours daily over a period of many years. Excessive exposure to mechanical vibrations often causes low back pain and musculoskeletal problems, and may harm some organs in the human body. In this way, the present research includes the monitoring of four data collection points, considering the front and rear axles of a sprayer, the operator cabin floor and the operator seat in a self-propelled sprayer. The vibration transmissibility between these points is used to measure the vibration severity to which the operator is exposed under different forward speeds and tire inflation pressure conditions. The RMS acceleration levels for both the cabin floor and the operator’s seat were classified as “uncomfortable” and “very uncomfortable” for a workload of 8 h according to the ISO 2631-1, which indicates that the vibration levels that affect the agricultural machinery operator should be reduced. The vibration transmissibility was greater than 1 when measured between the rear axle and the floor of the operating cabin. The vibration transmissibility from the floor to the seat was lower than 1 in all scenarios evaluated, which indicates that seat damping is effective since the vibration severity that affects the operator seat is lower than the vibration severity of the cabin floor. Full article

Recent advances in generative artificial intelligence (GenAI) have enabled the creation of high-resolution synthetic images, offering an alternative to traditional data collection for training computer vision models in agriculture. In crop disease diagnosis, synthetic images can supplement datasets when real image acquisition is limited, potentially reducing resource-intensive field collection. Therefore, this study evaluated how different ratios of real-field to Gen-AI-based synthetic watermelon (Citrullus lanatus) disease images (including an additional unknown class) affect EfficientNetV2-L classification performance and feature-space separability. The training dataset was divided into five treatments: H0 (real images only), H1 (synthetic images only), H2 (equal real-to-synthetic ratio), H3 (one real image to ten synthetic images, 1:10), and H4 (H3 plus random images to enhance variability). Models were trained using a custom EfficientNetV2-L architecture with fine-tuning and transfer learning approaches. Treatments H2, H3, and H4 demonstrated strong and consistent performance across all classes, with H2 achieving overall accuracy of 0.80, followed by H3 (0.98) and H4 (0.98). H3 achieved near-perfect precision and recall (0.95–0.99) across all classes, resulting in F1-scores of 0.98. H4 also maintained high precision and recall scores (0.94–1.00), including accurate detection of the additional unknown class (F1 = 0.98). Overall weighted F1-scores increased substantially from 0.72 (H0) to 0.81 (H2) and reached 0.98 in H3-H4, indicating the benefit of hybrid synthetic-real data fusion. These findings show that real-synthetic data fusion enhances model performance and generalization, while synthetic images alone were not effective under the tested conditions. Full article

Biochar application has been proposed as a promising strategy to improve soil functioning, defined as the integrated regulation of water storage, nutrient availability, and biological activity influencing crop productivity and crop performance in water-limited environments. However, its effectiveness depends on soil properties, climatic variability, and dominant processes. This study evaluated the effects of sewage sludge biochar on soil quality, water dynamics, nutrient availability, and bean productivity in sandy soil under rainfed semiarid conditions across two contrasting cropping cycles. A soil quality index (SQI) based on a minimum data set (MDS) derived from principal component analysis (PCA) was used to identify the dominant processes controlling soil functioning under different hydrological regimes. The two cropping cycles corresponded to wetter (Cycle I) and drier (Cycle II) hydrological conditions within the same agricultural year. Biochar application increased soil organic carbon and nitrogen stocks, enhanced phosphorus availability, and improved soil water storage. Despite similar evapotranspiration among treatments, water productivity increased, indicating more efficient conversion of stored soil water into yield. Biological indicators were more responsive during the wetter cycle, whereas physicochemical indicators dominated under drier conditions, revealing a shift in the processes regulating soil functioning. The minimum data set varied between cycles, demonstrating the environmental dependency of the SQI components. Overall, biochar improved soil resilience by enhancing nutrient retention and buffering crop response to water limitation, and the integrative SQI approach effectively captured these functional changes. Full article

Destemming is a critical operation in winemaking, determining the proportion of stems, skins, and vegetal fragments that enter the must and influencing both the mechanical integrity of berries and the extraction kinetics during fermentation. Horizontal centrifugal destemmers, which rely on a rotating beater and a stationary perforated cage, represent the most widely adopted technology in modern wineries. Despite their prevalence, the mechanical basis of berry detachment remains poorly quantified, with operational guidelines grounded largely in empirical practices rather than mechanistic understanding. This study develops an extended theoretical–numerical framework describing the forces acting on grape berries during destemming, focusing on the contributions of centrifugal force, tangential shear, and impact loading. Using realistic machine geometry and published distributions of pedicel detachment strength, the governing equations for each force regime are derived, and their interaction is evaluated through Monte Carlo simulations (n = 10,000). Results demonstrate that tangential shear is the dominant mechanism of detachment, particularly at lower rotational speeds where torque is maximized according to mechanical transmission principles. Centrifugal forces contribute modestly, while impact forces are largely ineffective for detachment and are instead associated with berry damage. The work provides a comprehensive reinterpretation of destemming mechanics and offers new guidance for machine design and operational strategies aimed at improving detachment efficiency while reducing berry damage. Full article

Outbreaks of stored grain pests can pose significant threats to food security. In-depth analyses of sudden outbreaks are key to achieving effective prevention and control. To address the issue of models’ insufficient reasoning capability arising from complex causal relationships in stored grain pest events, this study proposes an Event Patterns Enhancing Causal Reasoning (EPECR) method incorporating category theory. Specifically, we focus on common pests such as Sitophilus zeamais (maize weevil) and Sitotroga cerealella (Angoumois grain moth). We formally map the domain ontology—including entities like environmental factors (e.g., temperature, humidity) and control measures (e.g., fumigation)—to categories, and represent their inter-relationships (e.g., inhibition, promotion) as functors. To handle complex scenarios, we model multi-cause events (e.g., high temperature and humidity jointly accelerating pest reproduction) using functor products, and represent multi-hop events (e.g., environmental changes leading to pest outbreak and subsequent grain loss) through functor compositions. This formal expression enables Large Language Models (LLMs) to extract reliable event patterns. Based on these patterns, this study constructed 1440 structured datasets and adopted the Low-Rank Adaptation (LoRA) strategy to fine-tune the LLMs. Experiments on the domain-specific Stored Grain Pest Events Dataset (SGPE) demonstrate that EPECR achieves a reasoning accuracy of 85.9% on in-distribution data and 79.9% on out-of-distribution data, effectively identifying correct causal chains for pest logic. This method significantly outperforms the state-of-the-art domain method-Naive Augmentations (NA)-by 4.9%, providing precise decision support for the early warning and control of specific pest incidents. Full article

To address persistent problems such as clogging, high digging resistance, incomplete soil removal, and severe pod loss during the operation of shovel-chain peanut harvesters, a hybrid excavation approach was developed based on an in-depth analysis of the mechanical interaction between the peanut plant–soil complex (hereafter referred to as the “complex”) and the harvesting mechanism. The proposed approach integrates vertical and horizontal excavation directions to enhance soil fragmentation and reduce operational resistance. A progressive soil disintegration process was introduced, in which the complex undergoes lateral and longitudinal compression-bending deformation during movement. A driven soil–plant separation scheme was implemented through coordinated operation of upper conveying and lower combing–lifting mechanisms, promoting efficient and continuous material flow. A resistance-reducing digging device consisting of opposing round plow blades and horizontally sliding digging shovels was designed to minimize excavation resistance and soil adhesion. Meanwhile, an anti-clogging separation mechanism, integrating squeezing and feeding rollers and harrow-chain, was developed to improve soil removal and pod separation. Key structural and operational parameters—such as the chain-to-machine speed ratio, tooth-to-chain rotation speed ratio, harrow-tooth spacing ratio, and pushing-tooth transmission ratio—were optimized through theoretical analysis and prototyping. The final design also refined the number of pushing-tooth rows, squeezing and feeding roller geometry, conveying-tooth radius, and the configuration and distribution of rake and stick-tooth shafts. Field experiments were conducted using the developed prototype under sandy loam conditions (11–15% moisture content) with Yu Hua 22 peanut plants (35–40 cm height, 70 cm ridge spacing, 30 cm narrow-row spacing) at a working speed of 1.5–1.6 km·h⁻¹. Results demonstrated that the prototype achieved average ground pod loss, buried pod, and soil carryover rates of 1.13%, 0.95%, and 7.87%, respectively. The entire operation proceeded smoothly without clogging, and continuous conveying of peanut plants was maintained. These findings confirm that the proposed combined excavation and separation system meets and in some respects exceeds the performance requirements for efficient peanut harvesting under typical field conditions. Full article

The optimization of agrochemical spraying can be approached by increasing the efficiency of product distribution, which improves application quality and the biological effectiveness of the treatment. This study presents the development and evaluation of four distinct control strategies to adjust the hydraulic system of a new small, low-cost, electric, vertical variable-rate sprayer based on variations in travel speed, aiming to maintain a constant spray volume during operation and, thereby, increase distribution efficiency. The evaluated algorithms were developed from a mathematical model of the prototype’s hydraulic system obtained from experimental data and using the system identification tool in MATLAB software version 2021. Two open-loop algorithms (linear regression and Fuzzy) and two closed-loop algorithms (Integral and Fuzzy-PD with output integration) were developed. The evaluation was conducted through simulations, using a normalized speed data series provided by the United States Environmental Protection Agency. Performance evaluation results determined that the Fuzzy-PD algorithm with output integration showed the best performance (ISE = 0.21 × 10⁻⁵), followed by the linear regression algorithm (ISE = 3.38 × 10⁻⁵). The results demonstrated that, compared to applications based on fixed rates defined by nominal parameters, the developed sprayer showed potential to improve the uniformity of spray distribution in the field. Full article

Advanced sensing technologies increasingly support monitoring and decision-making processes in modern agriculture. This study investigates the feasibility of developing a harvest timing monitoring workflow based on a portable hyperspectral imaging (HSI) system in the visible–near-infrared (VIS-NIR: 400–1000 nm) range, coupled with machine learning. A hierarchical Partial Least Squares–Discriminant Analysis (Hi-PLS-DA) model was developed and tested to discriminate harvestable from non-harvestable plants of Brassica rapa subsp. sylvestris through the identification of open flowers within otherwise closed flower buds in the raceme. The classification included four target plant classes, i.e., green inflorescences, green leaves, yellow flowers, and yellow leaves, along with two non-target classes, background and not-classified (NC), which were included to support the classification process. The predicted hyperspectral images demonstrated a clear distinction between closed and open flowers, supported by satisfactory classification performance (sensitivity, specificity, precision, and F1-score: 0.78–1.00). This workflow proved effective in handling intrinsic outdoor hyperspectral variability, mitigating illumination and canopy texture, and offers useful methodological insights for the possible future integration of HSI-based approaches into automated field applications, paving the way for rapid, real-time harvest decision support. Full article