We investigate the impact of domain-specific self-supervised pre-training on agricultural disease classification using hierarchical vision transformers. Our key finding is that SimCLR pre-training on just 3,000 unlabeled agricultural images provides a +4.57% accuracy improvement--exceeding the +3.70% gain from hierarchical architecture design. Critically, we show this SSL benefit is architecture-agnostic: applying the same pre-training to Swin-Base yields +4.08%, to ViT-Base +4.20%, confirming practitioners should prioritize domain data collection over architectural choices. Using HierarchicalViT (HVT), a Swin-style hierarchical transformer, we evaluate on three datasets: Cotton Leaf Disease (7 classes, 90.24%), PlantVillage (38 classes, 96.3%), and PlantDoc (27 classes, 87.1%). At matched parameter counts, HVT-Base (78M) achieves 88.91% vs. Swin-Base (88M) at 87.23%, a +1.68% improvement. For deployment reliability, we report calibration analysis showing HVT achieves 3.56% ECE (1.52% after temperature scaling). Code: https://github.com/w2sg-arnav/HierarchicalViT
Accurate reconstruction of leaf surfaces from 3D point cloud is essential for agricultural applications such as phenotyping. However, real-world plant data (i.e., irregular 3D point cloud) are often complex to reconstruct plant parts accurately. A wide range of surface reconstruction methods has been proposed, including parametric, triangulation-based, implicit, and learning based approaches, yet their relative performance for leaf surface reconstruction remains insufficiently understood. In this work, we present a comparative study of nine representative surface reconstruction methods for leaf surfaces. We evaluate these methods on three publicly available datasets: LAST-STRAW, Pheno4D, and Crops3D - spanning diverse species, sensors, and sensing environments, ranging from clean high-resolution indoor scans to noisy low-resolution field settings. The analysis highlights the trade-offs between surface area estimation accuracy, smoothness, robustness to noise and missing data, and computational cost across different methods. These factors affect the cost and constraints of robotic hardware used in agricultural applications. Our results show that each method exhibits distinct advantages depending on application and resource constraints. The findings provide practical guidance for selecting surface reconstruction techniques for resource constrained robotic platforms.
Abstract Agricultural sectors in fragile and post-conflict countries often face fragmented knowledge systems, weak institutional learning, and limited access to innovative practices, constraining productivity, employment, and market performance. This study investigates how knowledge management (KM) maturity influences agricultural economic development in Iraq, integrating qualitative and quantitative evidence. A sequential mixed-methods design was implemented. Phase 1 involved directed content analysis of 88 publications (2007–2022) to identify four core KM components: Knowledge Acquisition, Knowledge Storage, Knowledge Distribution, and Knowledge Application. Phase 2 employed semi-structured interviews with 24 agricultural and KM experts to contextualize the framework and define four outcome constructs: Infrastructure Investment, Employment, Trade and Market Performance, and Productivity and Profitability. Phase 3 validated a 70-item survey using CVR and CVI. Phase 4 applied the survey to 261 employees of the Najaf Agricultural Directorate, with analysis using PLS-SEM4 and CB-SEM (AMOS26). Results indicate that Knowledge Acquisition (Mean = 3.38) and Knowledge Application (Mean = 3.28) are the most developed components, while Knowledge Storage (Mean = 2.80) remains a bottleneck. SEM analysis confirms that KM maturity significantly drives economic development (β = 0.66, p < 0.001), explaining 44% of variance. Trade and Market exhibits the strongest structural effect, followed by Employment and Productivity and Profitability; Infrastructure Investment shows a weaker effect. Knowledge Acquisition and Application primarily enhance productivity and employment, whereas Storage and Transfer are critical for market efficiency and institutional learning. These findings underscore that strengthening KM, particularly in applying and transferring knowledge, is pivotal for sustainable agricultural development in fragile contexts. The study provides an empirically validated framework for designing targeted KM interventions to maximize productivity, market integration, and economic resilience.
Relevance. In the industrially tense region (Donbass), as a result of socio-economic upheavals since 2014, many lands have been withdrawn from agricultural use and are now abandoned and degrading. Areas of active military action create beligerative landscapes characterized by profound geophysical and geochemical transformations. These areas are hotbeds of toxic environmental impacts and require targeted restoration measures. Phytoremediation stands out among the most effective methods for optimizing natural-territorial complexes of the DPR as the most effective, economically advantageous and aesthetically attractive.Materials and Methods. Agricultural and recreational ecotopes in the Central Donbass were studied. A field assessment of the state of local geosystems was conducted. Morphological analysis and description of plants, as well as calculations for determining life strategies (CSR), were applied. Analytical methods (atomic absorption, inductively coupled plasma mass spectrometry, and neutron activation) were used.Results. A difference in the range of informative structural features variation of some indicator plants for use in phytoremediation purposes in post-conflict areas – sites of active military operations in Donbass – has been established. New geochemical anomalies were identified in post-conflict areas for a number of technophile elements (Mn, Р, Zn, Cu, Mo, Ni, Pb, Cr, La, Co, Se, As, Cd). For the plant species Cichorium intybus L., Taraxacum officinale F.H.Wigg, Plantago major L., and Diplotaxis muralis (L.) DC., the implementation patterns of life-sustaining strategies (visualization of CSR in the Grime-Ramensky triangle) and ecological plasticity in areas affected by the militarization of the region were determined. Anatomical and morphological pathologies of the studied species were identified. The ecological valence of species allows them to support the initial stages of active succession during the first two to three years, forming a vegetation cover that performs anti-erosion and habitatforming functions. Based on plant morphopathologies and elemental composition data, geochemical anomalies were identified and a range of geochemical background values for elemental composition in plant samples was described. A phosphorus-lanthanum anomaly (P-La), a consequence of military operations in the DPR, is described for the first time.
Leonardo Grando, Juan Fernando Galindo Jaramillo, Jose Roberto Emiliano Leite
et al.
The low battery autonomy of Unnamed Aerial Vehicles (UAVs or drones) can make smart farming (precision agriculture), disaster recovery, and the fighting against dengue vector applications difficult. This article considers two approaches, first enumerating the characteristics observed in these three IoT application types and then modeling an UAV's battery recharge coordination using the Agent-Based Simulation (ABS) approach. In this way, we propose that each drone inside the swarm does not communicate concerning this recharge coordination decision, reducing energy usage and permitting remote usage. A total of 6000 simulations were run to evaluate how two proposed policies, the BaseLine (BL) and ChargerThershold (CT) coordination recharging policy, behave in 30 situations regarding how each simulation sets conclude the simulation runs and how much time they work until recharging results. CT policy shows more reliable results in extreme system usage. This work conclusion presents the potential of these three IoT applications to achieve their perpetual service without communication between drones and ground stations. This work can be a baseline for future policies and simulation parameter enhancements.
Alberto San-Miguel-Tello, Gennaro Scarati, Alejandro Hernández
et al.
This paper presents advances on the Universal Manipulation Interface (UMI), a low-cost hand-held gripper for robot Learning from Demonstration (LfD), for complex in-the-wild scenarios found in agricultural settings. The focus is on improving the acquisition of suitable samples with minimal additional setup. Firstly, idle times and user's cognitive load are reduced through the extraction of individual samples from a continuous demonstration considering task events. Secondly, reliability on the generation of task sample's trajectories is increased through the combination on-board inertial measurements and external visual marker localization usage using Extended Kalman Filtering (EKF). Results are presented for a fruit harvesting task, outperforming the default pipeline.
Biotic pest attacks and infestations are major causes of stored grain losses, leading to significant food and economic losses. Conventional, manual, sampling-based pest recognition methods are labor-intensive, time-consuming, costly, require expertise, and may not even detect hidden infestations. In recent years, the electronic nose (e-nose) approach has emerged as a potential alternative for agricultural grain pest recognition and monitoring. An e-nose mimics human olfactory systems by integrating a sensor array, data acquisition, and analysis for recognizing grain pests by analyzing volatile organic compounds (VOCs) emitted by grain and pests. However, well-documented, curated, and synthesized literature on the use of e-nose technology for grain pest detection is lacking. Therefore, this systematic literature review provides a comprehensive overview of the current state-of-the-art e-nose technology for agricultural grain pest monitoring. The review examines employed sensor technology, targeted pest species type, grain medium, data processing, and pattern recognition techniques. An e-nose is a promising tool that offers a rapid, low-cost, non-destructive solution for detecting, identifying, and monitoring grain pests, including microscopic and hidden insects, with good accuracy. We identified the factors that influence the e-nose performance, which include pest species, storage duration, temperature, moisture content, and pest density. The major challenges include sensor array optimization or selection, large data processing, poor repeatability, and comparability among measurements. An inexpensive and portable e-nose has the potential to help stakeholders and storage managers take timely and data-driven informed actions or decisions to reduce overall food and economic losses.
Litchi is a high-value fruit, yet traditional manual selection methods are increasingly inadequate for modern production demands. Integrating UAV-based aerial imagery with deep learning offers a promising solution to enhance efficiency and reduce costs. This paper introduces YOLOv11-Litchi, a lightweight and robust detection model specifically designed for UAV-based litchi detection. Built upon the YOLOv11 framework, the proposed model addresses key challenges such as small target size, large model parameters hindering deployment, and frequent target occlusion. To tackle these issues, three major innovations are incorporated: a multi-scale residual module to improve contextual feature extraction across scales, a lightweight feature fusion method to reduce model size and computational costs while maintaining high accuracy, and a litchi occlusion detection head to mitigate occlusion effects by emphasizing target regions and suppressing background interference. Experimental results validate the model's effectiveness. YOLOv11-Litchi achieves a parameter size of 6.35 MB - 32.5% smaller than the YOLOv11 baseline - while improving mAP by 2.5% to 90.1% and F1-Score by 1.4% to 85.5%. Additionally, the model achieves a frame rate of 57.2 FPS, meeting real-time detection requirements. These findings demonstrate the suitability of YOLOv11-Litchi for UAV-based litchi detection in complex orchard environments, showcasing its potential for broader applications in precision agriculture.
Accurate crop mapping fundamentally relies on modeling multi-scale spatiotemporal patterns, where spatial scales range from individual field textures to landscape-level context, and temporal scales capture both short-term phenological transitions and full growing-season dynamics. Transformer-based remote sensing foundation models (RSFMs) offer promising potential for crop mapping due to their innate ability for unified spatiotemporal processing. However, current RSFMs remain suboptimal for crop mapping: they either employ fixed spatiotemporal windows that ignore the multi-scale nature of crop systems or completely disregard temporal information by focusing solely on spatial patterns. To bridge these gaps, we present AgriFM, a multi-source remote sensing foundation model specifically designed for agricultural crop mapping. Our approach begins by establishing the necessity of simultaneous hierarchical spatiotemporal feature extraction, leading to the development of a modified Video Swin Transformer architecture where temporal down-sampling is synchronized with spatial scaling operations. This modified backbone enables efficient unified processing of long time-series satellite inputs. AgriFM leverages temporally rich data streams from three satellite sources including MODIS, Landsat-8/9 and Sentinel-2, and is pre-trained on a global representative dataset comprising over 25 million image samples supervised by land cover products. The resulting framework incorporates a versatile decoder architecture that dynamically fuses these learned spatiotemporal representations, supporting diverse downstream tasks. Comprehensive evaluations demonstrate AgriFM's superior performance over conventional deep learning approaches and state-of-the-art general-purpose RSFMs across all downstream tasks. Codes will be available at https://github.com/flyakon/AgriFM.
There is a growing demand for autonomous mobile robots capable of navigating unstructured agricultural environments. Tasks such as weed control in meadows require efficient path planning through an unordered set of coordinates while minimizing travel distance and adhering to curvature constraints to prevent soil damage and protect vegetation. This paper presents an integrated navigation framework combining a global path planner based on the Dubins Traveling Salesman Problem (DTSP) with a Nonlinear Model Predictive Control (NMPC) strategy for local path planning and control. The DTSP generates a minimum-length, curvature-constrained path that efficiently visits all targets, while the NMPC leverages this path to compute control signals to accurately reach each waypoint. The system's performance was validated through comparative simulation analysis on real-world field datasets, demonstrating that the coupled DTSP-based planner produced smoother and shorter paths, with a reduction of about 16% in the provided scenario, compared to decoupled methods. Based thereon, the NMPC controller effectively steered the robot to the desired waypoints, while locally optimizing the trajectory and ensuring adherence to constraints. These findings demonstrate the potential of the proposed framework for efficient autonomous navigation in agricultural environments.
Accurate and robust environmental perception is crucial for robot autonomous navigation. While current methods typically adopt optical sensors (e.g., camera, LiDAR) as primary sensing modalities, their susceptibility to visual occlusion often leads to degraded performance or complete system failure. In this paper, we focus on agricultural scenarios where robots are exposed to the risk of onboard sensor contamination. Leveraging radar's strong penetration capability, we introduce a radar-based 3D environmental perception framework as a viable alternative. It comprises three core modules designed for dense and accurate semantic perception: 1) Parallel frame accumulation to enhance signal-to-noise ratio of radar raw data. 2) A diffusion model-based hierarchical learning framework that first filters radar sidelobe artifacts then generates fine-grained 3D semantic point clouds. 3) A specifically designed sparse 3D network optimized for processing large-scale radar raw data. We conducted extensive benchmark comparisons and experimental evaluations on a self-built dataset collected in real-world agricultural field scenes. Results demonstrate that our method achieves superior structural and semantic prediction performance compared to existing methods, while simultaneously reducing computational and memory costs by 51.3% and 27.5%, respectively. Furthermore, our approach achieves complete reconstruction and accurate classification of thin structures such as poles and wires-which existing methods struggle to perceive-highlighting its potential for dense and accurate 3D radar perception.
Mangesh Vaidya, V. R. Patodkar, Prajakta Kuralkar
et al.
Livestock-generated methane, particularly from cattle, was a significant contributor to climate change. Methane emissions from ruminant animals, such as cows and sheep, are primarily caused by the microbial fermentation of food in their digestive systems, a process known as enteric fermentation by making this process a prime source of greenhouse gas emissions in animal production. Considerable knowledge gaps existed in animal agriculture regarding effective strategies for mitigating these emissions while maintaining productivity. A key factor was the uncertainty surrounding methods for estimating emission rates, each having inherent limitations. For example, the suitability of the GreenFeed system varied based on specific experiment objectives. Compared to respiration chambers and the sulfur hexafluoride tracer method, the The GreenFeed system often required more time and a larger number of animals for treatment comparisons due to higher within-day variances. It measured numerous short-term methane emissions from individual animals at various times throughout the day over several days. Recent advancements focused on improving accuracy, ease of use, and cost-effectiveness, essential for better monitoring of greenhouse gases. Traditional methods, such as respiration chambers, while accurate, were costly and impractical for field measurements. The GreenFeed system’s software facilitated control over feed availability timing and CH4 measurement allocation. Therefore, careful planning was necessary to ensure accurate estimates of methane production. This review emphasized the need for effective measurement techniques to mitigate methane emissions from livestock.
Kevin A. Adkins, Kevin Li, Maximilian N. Blasko
et al.
Abstract Context Understanding the movement of bioaerosols, such as spores and pollen, through the atmosphere is important for a broad spectrum of landscape research, including agricultural fungal outbreaks and pollen threats to public health. As spores and pollen can be transported in the air over large distances, the use of aircraft has historically played a role in detecting and mapping their presence in the lower atmosphere. Objectives We present a simple alternative to costly and specialized aircraft and associated equipment that are typically used in the study of spores and pollen in the atmosphere. Methods We use 3D printable components and common lab supplies mounted on an uncrewed aircraft (UA). Conveniently, this setup does not require additional electronic components to control collection during flight, using the UA landing gear mechanism instead. Results We demonstrate that this apparatus can collect fungal spores in the atmosphere and describe potential impacts by the environment and experimental protocol on collection efficiency. These include the effects of: (1) competing airflows from UA rotors, flight trajectories, and wind, (2) flight altitude, and (3) particle size and Petri dish collection medium. Conclusions Complex biological mechanisms and atmospheric dynamics dictate the release, transport, and deposition of bioaerosols. Economical methods to sample bioaerosols in the lower atmosphere can increase the amount and type of data collected and unlock new understanding. The methodology presented here provides an economical method to sample bioaerosols that can help improve landscape-level understanding of the dispersal of bioaerosols.
Agricultural management, with a particular focus on fertilization strategies, holds a central role in shaping crop yield, economic profitability, and environmental sustainability. While conventional guidelines offer valuable insights, their efficacy diminishes when confronted with extreme weather conditions, such as heatwaves and droughts. In this study, we introduce an innovative framework that integrates Deep Reinforcement Learning (DRL) with Recurrent Neural Networks (RNNs). Leveraging the Gym-DSSAT simulator, we train an intelligent agent to master optimal nitrogen fertilization management. Through a series of simulation experiments conducted on corn crops in Iowa, we compare Partially Observable Markov Decision Process (POMDP) models with Markov Decision Process (MDP) models. Our research underscores the advantages of utilizing sequential observations in developing more efficient nitrogen input policies. Additionally, we explore the impact of climate variability, particularly during extreme weather events, on agricultural outcomes and management. Our findings demonstrate the adaptability of fertilization policies to varying climate conditions. Notably, a fixed policy exhibits resilience in the face of minor climate fluctuations, leading to commendable corn yields, cost-effectiveness, and environmental conservation. However, our study illuminates the need for agent retraining to acquire new optimal policies under extreme weather events. This research charts a promising course toward adaptable fertilization strategies that can seamlessly align with dynamic climate scenarios, ultimately contributing to the optimization of crop management practices.
Arun N. Sivakumar, Mateus V. Gasparino, Michael McGuire
et al.
We present a vision-based navigation system for under-canopy agricultural robots using semantic keypoints. Autonomous under-canopy navigation is challenging due to the tight spacing between the crop rows ($\sim 0.75$ m), degradation in RTK-GPS accuracy due to multipath error, and noise in LiDAR measurements from the excessive clutter. Our system, CropFollow++, introduces modular and interpretable perception architecture with a learned semantic keypoint representation. We deployed CropFollow++ in multiple under-canopy cover crop planting robots on a large scale (25 km in total) in various field conditions and we discuss the key lessons learned from this.
As the global food system faces increasing challenges from sustainability, climate change, and food security issues, alternative food networks like Community-Supported Agriculture (CSA) play an essential role in fostering stronger connections between consumers and producers. However, understanding consumer engagement with CSA is fragmented, particularly in Japan where CSA participation is still emerging. This study aims to identify potential CSA participants in Japan and validate existing theories on CSA participation through a quantitative analysis of 2,484 Japanese consumers. Using choice experiments, Latent Class Analysis, and Partial Least Squares Structural Equation Modeling, we identified five distinct consumer segments. The "Sustainable Food Seekers" group showed the highest positive utility for CSA, driven primarily by "Food Education and Learning Opportunities" and "Contribution to Environmental and Social Issues." These factors were consistently significant across all segments, suggesting that many Japanese consumers value CSA for its educational and environmental benefits. In contrast, factors related to "Variety of Ingredients" were less influential in determining participation intentions. The findings suggest that promoting CSA in Japan may be most effective by emphasizing its role in environmental and social impact, rather than focusing solely on product attributes like organic certification, which is readily available in supermarkets. This reflects a key distinction between CSA adoption in Japan and in other cultural contexts, where access to organic produce is a primary driver. For "Sustainable Food Seekers," CSA offers a way to contribute to broader societal goals rather than just securing organic products.
Md. Toukir Ahmed, Arthur Villordon, Mohammed Kamruzzaman
Hyperspectral imaging (HSI) has become a key technology for non-invasive quality evaluation in various fields, offering detailed insights through spatial and spectral data. Despite its efficacy, the complexity and high cost of HSI systems have hindered their widespread adoption. This study addressed these challenges by exploring deep learning-based hyperspectral image reconstruction from RGB (Red, Green, Blue) images, particularly for agricultural products. Specifically, different hyperspectral reconstruction algorithms, such as Hyperspectral Convolutional Neural Network - Dense (HSCNN-D), High-Resolution Network (HRNET), and Multi-Scale Transformer Plus Plus (MST++), were compared to assess the dry matter content of sweet potatoes. Among the tested reconstruction methods, HRNET demonstrated superior performance, achieving the lowest mean relative absolute error (MRAE) of 0.07, root mean square error (RMSE) of 0.03, and the highest peak signal-to-noise ratio (PSNR) of 32.28 decibels (dB). Some key features were selected using the genetic algorithm (GA), and their importance was interpreted using explainable artificial intelligence (XAI). Partial least squares regression (PLSR) models were developed using the RGB, reconstructed, and ground truth (GT) data. The visual and spectra quality of these reconstructed methods was compared with GT data, and predicted maps were generated. The results revealed the prospect of deep learning-based hyperspectral image reconstruction as a cost-effective and efficient quality assessment tool for agricultural and biological applications.
We investigated the effect of feed deprivation for 45 days on the growth, immunity, and health of 0<sup>+</sup>marron (<i>Cherax cainii</i>) initially fed for 110 days on various protein sources including fishmeal (FM), poultry by-product meal (PBM), black soldier fly meal (BSFM), soybean meal (SBM), lupin meal (LM), and tuna hydrolysate. The marron were weighed and sacrificed immediately after feeding stopped (day 0) and at days 15, 30, and 45 after the feed deprivation trial commenced. Total haemolymph count, differential haemocyte count, lysozyme activity, protease activity, total bacterial count in the digestive tract, and organosomatic indices were analysed. Initially feeding marron any protein sources did not influence the percentage of weight gain and specific growth rates of marron. All marron showed more than 83% survival; however, marron fed soybean meal showed significantly lower survival than others. Dietary sources of protein altered organosomatic indices of starved marron during various starvation periods and resulted in a significant decrease in total haemocyte counts, lysozyme activity, protease activity, and bacterial count in the digestive tract of marron. Starved marron initially fed PBM and BSFM showed higher tolerance to starvation, followed by marron initially fed FM and SBM, while marron initially fed TH and LM showed the highest susceptibility to starvation.
Sustainable intensification calls for agroecological and adaptive management of the agrifood system. Here, we focus on intercropping and how this agroecological practice can be used to increase the sustainability of crop production. Strip, mixed, and relay intercropping can be used to increase crop yields through resource partitioning and facilitation. In addition to achieving greater productivity, diversifying cropping systems through the use of strategic intercrops can increase yield stability, reduce pests, and improve soil health. Several intercropping systems are already implemented in industrialized agricultural landscapes, including mixed intercropping with perennial grasses and legumes as forage and relay intercropping with winter wheat and red clover. Because intercropping can provide numerous benefits, researchers should be clear about their objectives and use appropriate methods so as to not draw spurious conclusions when studying intercrops. In order to advance the practice, experiments that test the effects of intercropping should use standardized methodology, and researchers should report a set of common criteria to facilitate cross-study comparisons. Intercropping with two or more crops appears to be less common with annuals than perennials, which is likely due to differences in the mechanisms responsible for complementarity. One area where intercropping with annuals in industrialized agricultural landscapes has advanced is with cover crops, where private, public, and governmental organizations have harmonized efforts to increase the adoption of cover crop mixtures.
This paper presents an approach based on higher order dynamic mode decomposition (HODMD) to model, analyse, and forecast energy behaviour in an urban agriculture farm situated in a retrofitted London underground tunnel, where observed measurements are influenced by noisy and occasionally transient conditions. HODMD is a data-driven reduced order modelling method typically used to analyse and predict highly noisy and complex flows in fluid dynamics or any type of complex data from dynamical systems. HODMD is a recent extension of the classical dynamic mode decomposition method (DMD), customised to handle scenarios where the spectral complexity underlying the measurement data is higher than its spatial complexity, such as is the environmental behaviour of the farm. HODMD decomposes temporal data as a linear expansion of physically-meaningful DMD-modes in a semi-automatic approach, using a time-delay embedded approach. We apply HODMD to three seasonal scenarios using real data measured by sensors located at at the cross-sectional centre of the the underground farm. Through the study we revealed three physically-interpretable mode pairs that govern the environmental behaviour at the centre of the farm, consistently across environmental scenarios. Subsequently, we demonstrate how we can reconstruct the fundamental structure of the observed time-series using only these modes, and forecast for three days ahead, as one, compact and interpretable reduced-order model. We find HODMD to serve as a robust, semi-automatic modelling alternative for predictive modelling in Digital Twins.