Vision Foundation Models (VFM) pre-trained on large-scale unlabeled data have achieved remarkable success on general computer vision tasks, yet typically suffer from significant domain gaps when applied to agriculture. In this context, we introduce $SPROUT$ ($S$calable $P$lant $R$epresentation model via $O$pen-field $U$nsupervised $T$raining), a multi-crop, multi-task agricultural foundation model trained via diffusion denoising. SPROUT leverages a VAE-free Pixel-space Diffusion Transformer to learn rich, structure-aware representations through denoising and enabling efficient end-to-end training. We pre-train SPROUT on a curated dataset of 2.6 million high-quality agricultural images spanning diverse crops, growth stages, and environments. Extensive experiments demonstrate that SPROUT consistently outperforms state-of-the-art web-pretrained and agricultural foundation models across a wide range of downstream tasks, while requiring substantially lower pre-training cost. The code and model are available at https://github.com/UTokyo-FieldPhenomics-Lab/SPROUT.
Despite rapid advances in multimodal large language models, agricultural applications remain constrained by the lack of multilingual speech data, unified multimodal architectures, and comprehensive evaluation benchmarks. To address these challenges, we present AgriGPT-Omni, an agricultural omni-framework that integrates speech, vision, and text in a unified framework. First, we construct a scalable data synthesis and collection pipeline that converts agricultural texts and images into training data, resulting in the largest agricultural speech dataset to date, including 492K synthetic and 1.4K real speech samples across six languages. Second, based on this, we train the first agricultural omni-model via a three-stage paradigm: textual knowledge injection, progressive multimodal alignment, and GRPO-based reinforcement learning, enabling unified reasoning across languages and modalities. Third, we propose AgriBench-Omni-2K, the first tri-modal benchmark for agriculture, covering diverse speech-vision-text tasks and multilingual slices, with standardized protocols and reproducible tools. Experiments show that AgriGPT-Omni significantly outperforms general-purpose baselines on multilingual and multimodal reasoning as well as real-world speech understanding. All models, data, benchmarks, and code will be released to promote reproducible research, inclusive agricultural intelligence, and sustainable AI development for low-resource regions.
Purpose: The study examines the Open Access (OA) landscape of Indian state agricultural universities, focusing on OA growth, leading institutions, prolific authors, preferred sources, funding, APC usage, and trending topics. It aims to identify research gaps, guide future research, and support policymakers in developing effective OA policies Design/methodology/approach The experiment utilized the OpenAlex database to collect global open access (OA) publications from Indian state agricultural universities over the past ten years (2014-2023). Using the Research Organization Registry ID, 97,536 publications were extracted. Data analysis was performed with OpenRefine, and ArcGIS 10.8 and Microsoft Excel were used for visualization. Findings: The global OA research output from state agricultural universities amounted to 65,889 publications across five OA categories: Green OA (7.35%), Diamond OA (6.74%), Gold OA (57.27%), Hybrid OA (9.24%), and Bronze OA (19.41%). Notably, 78.34% of articles were published in 864 low-impact domestic journals. Tamil Nadu Agricultural University produced the most publications in Gold, Diamond, Hybrid, and Bronze OA categories, while Punjab Agricultural University excelled in Green OA and received the highest funding, incurring the most article processing charges (APCs). Collaborative research focusing on agricultural policies, rice water management, soil fertility, and crop productivity had a greater impact. Originality/value The experiment is the first effort to evaluate the OA global academic research outputs of Indian state agriculture universities. The findings offer institutions, state governments, and funding agencies the opportunity to prioritise open-access publishing to promote sustainable agricultural research. Research limitations/implications The study is limited to the publications data indexed in the OpenAlex database.
Multi-sensor Simultaneous Localization and Mapping (SLAM) is essential for Unmanned Aerial Vehicles (UAVs) performing agricultural tasks such as spraying, surveying, and inspection. However, real-world, multi-modal agricultural UAV datasets that enable research on robust operation remain scarce. To address this gap, we present AgriLiRa4D, a multi-modal UAV dataset designed for challenging outdoor agricultural environments. AgriLiRa4D spans three representative farmland types-flat, hilly, and terraced-and includes both boundary and coverage operation modes, resulting in six flight sequence groups. The dataset provides high-accuracy ground-truth trajectories from a Fiber Optic Inertial Navigation System with Real-Time Kinematic capability (FINS_RTK), along with synchronized measurements from a 3D LiDAR, a 4D Radar, and an Inertial Measurement Unit (IMU), accompanied by complete intrinsic and extrinsic calibrations. Leveraging its comprehensive sensor suite and diverse real-world scenarios, AgriLiRa4D supports diverse SLAM and localization studies and enables rigorous robustness evaluation against low-texture crops, repetitive patterns, dynamic vegetation, and other challenges of real agricultural environments. To further demonstrate its utility, we benchmark four state-of-the-art multi-sensor SLAM algorithms across different sensor combinations, highlighting the difficulty of the proposed sequences and the necessity of multi-modal approaches for reliable UAV localization. By filling a critical gap in agricultural SLAM datasets, AgriLiRa4D provides a valuable benchmark for the research community and contributes to advancing autonomous navigation technologies for agricultural UAVs. The dataset can be downloaded from: https://zhan994.github.io/AgriLiRa4D.
Rishemjit Kaur, Arshdeep Singh Bhankhar, Jashanpreet Singh Salh
et al.
Enabling farmers to access accurate agriculture-related information in their native languages in a timely manner is crucial for the success of the agriculture field. Publicly available general-purpose Large Language Models (LLMs) typically offer generic agriculture advisories, lacking precision in local and multilingual contexts. Our study addresses this limitation by generating multilingual (English, Hindi, Punjabi) synthetic datasets from agriculture-specific documents from India and fine-tuning LLMs for the task of question answering (QA). Evaluation on human-created datasets demonstrates significant improvements in factuality, relevance, and agricultural consensus for the fine-tuned LLMs compared to the baseline counterparts.
Considering the status of water resources in the Urmia Lake basin (ULB) and the necessity for real-time monitoring of crop production and water consumption in this area, the objective of current research was to calculate actual evapotranspiration (ETa), actual crop coefficient (Kc act), water consumption, dry biomass, grain yield, and water productivity of irrigated wheat in the ULB using remote sensing techniques. For this purpose, the ULB was divided into six sub-basins. To calculate reference evapotranspiration (ETo), meteorological data and the FAO56 Penman-Monteith (P-M) equation were used, and ETa was calculated utilizing the Surface Energy Balance Algorithm for Land (SEBAL) and 43 Landsat 8 and 9 satellite images. Furthermore, to calculate biomass, grain yield, and crop water productivity (WPc), the Radiation Use Efficiency (RUE) model was employed. In this study, 1115 irrigated wheat farms were monitored. The results indicated that the average ETo in the ULB during the irrigated wheat growing season is 802 mm. Also, Kc act differed from FAO coefficients (Kc) across growth stages, being higher during initial and final stages but lower during development and intermediate stages. According to the results, the average water consumption volume for the entire basin was determined to be 4566 m3/ha. Also, the overall average Harvest Index (HI) for the ULB was found to be 0.42. The results of the biomass and grain yield investigation revealed significant differences in their values among different sub-basins. Additionally, the overall average WPc for irrigated wheat in the ULB was calculated to be 0.94 kg/m3.
Biochar and microorganisms are widely used soil amendments, but the effects of their combined application on crops and soils have not been thoroughly evaluated. We conducted a meta-analysis of 103 studies to examine the effects of biochar-microbe combined (BCM) application on crop physiology ecology and soil function. Together, they significantly improved crop growth and soil properties, with shoot dry weight increasing by 51.79 % and soil organic carbon increasing by 49.38 % compared with the control. BCM application can activate the antioxidant defense system in crops, reduce malondialdehyde levels by 18.24 %, and enhance soil bacterial abundance by 71.9 %. Their effect on soil pH was negatively correlated with the initial soil pH. The effects of BCM application vary with the properties of different biochar sources, pyrolysis temperatures, microbial species, and experiment types. Integrating microbes with low-temperature (≤ 500°C) wood-based biochar showed even better effects, not only enhancing crop growth performance and chlorophyll content, but also producing significant improvements in soil physicochemical properties. The combined application of different functional microorganisms and biochar resulted in varying effects on crop biomass accumulation, growth, and soil quality. In the study, BCM was evaluated for its ecological consequences on crops and soils, and we recommend prioritizing low-temperature biochar in combination with microorganisms to maximize crop and soil improvement.
Climate change nowadays became the most problematic matter including in agricultural industries. Agriculture area productivity affected national food security and a county’s economic development. As an agricultural county, Indonesia must be ready to adapt and prepare for the worst impact of climate change. This paper aims to explore the impact of climate change on financial sustainability in agricultural industries. This research uses a systematic literature review method related to financial sustainability, climate change impact, and agriculture industries. The result shows that financial sustainability in agricultural industries must be impacted by climate change. The impact of climate change on agriculture industries is associated with reducing profitability, destroying capital, portfolio reallocation, and financial instability. Climate change caused environmental uncertainty that affects agricultural productivity. To reduce the impact of climate change on financial sustainability in agricultural industries, there must be a design of mitigation must be prepared and realized so agricultural industries are more prepared and ready to face climate change impact.
Hanzhe Teng, Yipeng Wang, Dimitrios Chatziparaschis
et al.
Unmanned and intelligent agricultural systems are crucial for enhancing agricultural efficiency and for helping mitigate the effect of labor shortage. However, unlike urban environments, agricultural fields impose distinct and unique challenges on autonomous robotic systems, such as the unstructured and dynamic nature of the environment, the rough and uneven terrain, and the resulting non-smooth robot motion. To address these challenges, this work introduces an adaptive LiDAR odometry and mapping framework tailored for autonomous agricultural mobile robots operating in complex agricultural environments. The proposed framework consists of a robust LiDAR odometry algorithm based on dense Generalized-ICP scan matching, and an adaptive mapping module that considers motion stability and point cloud consistency for selective map updates. The key design principle of this framework is to prioritize the incremental consistency of the map by rejecting motion-distorted points and sparse dynamic objects, which in turn leads to high accuracy in odometry estimated from scan matching against the map. The effectiveness of the proposed method is validated via extensive evaluation against state-of-the-art methods on field datasets collected in real-world agricultural environments featuring various planting types, terrain types, and robot motion profiles. Results demonstrate that our method can achieve accurate odometry estimation and mapping results consistently and robustly across diverse agricultural settings, whereas other methods are sensitive to abrupt robot motion and accumulated drift in unstructured environments. Further, the computational efficiency of our method is competitive compared with other methods. The source code of the developed method and the associated field dataset are publicly available at https://github.com/UCR-Robotics/AG-LOAM.
Agricultural workers play a vital role in the global economy and food security by cultivating, transporting, and processing food for populations worldwide. Despite their importance, detailed spatial data on the global agricultural workforce have remained scarce. Here, we present a new gridded dataset that maps the global distribution of agricultural workers for every decade over the years 2000-2100, distributed at 0.083$\times$0.083 degrees resolution, roughly $\sim$10km$\times$10km at the Equator. The dataset is developed using an empirical modeling framework relying on generalized additive mixed models (GAMMs) that integrate socioeconomic variables, including gross domestic product per capita, total population, rural population size, and agricultural land use. The predictions are consistent with Shared Socio-economic Pathways and we distribute full time series data for all SSPs 1 to 5. This dataset opens new avenues for future research on labour force health, productivity and risk, and could be very useful for developing informed, forward-looking strategies that address the challenges of climate resilience in agriculture. The dataset and code for reproducing it are available for the user community [publicly available on publication at DOI: 10.5281/zenodo.14443333].
Agricultural products play a critical role in human development. With economic globalization and the financialization of agricultural products continuing to advance, the interconnections between different agricultural futures have become closer. We utilize a TVP-VAR-DY model combined with the quantile method to measure the risk spillover between 11 agricultural futures on the futures exchanges of US and China from July 9,2014, to December 31,2022. This study yielded several significant findings. Firstly, CBOT corn, soybean, and wheat were identified as the primary risk transmitters, with DCE corn and soybean as the main risk receivers. Secondly, sudden events or increased economic uncertainty can increase the overall risk spillovers. Thirdly, there is an aggregation of risk spillovers amongst agricultural futures based on the dynamic directional spillover results. Lastly, the central agricultural futures under the conditional mean are CBOT corn and soybean, while CZCE hard wheat and long-grained rice are the two risk spillover centers in extreme cases, as per the results of the spillover network and minimum spanning tree. Based on these results, decision-makers are advised to safeguard against the price risk of agricultural futures under sudden economic events, and investors can utilize the results to construct a superior investment portfolio by taking different agricultural product futures as risk-leading indicators according to various situations.
Vision is a major component in several digital technologies and tools used in agriculture. The object detector, You Look Only Once (YOLO), has gained popularity in agriculture in a relatively short span due to its state-of-the-art performance. YOLO offers real-time detection with good accuracy and is implemented in various agricultural tasks, including monitoring, surveillance, sensing, automation, and robotics. The research and application of YOLO in agriculture are accelerating rapidly but are fragmented and multidisciplinary. Moreover, the performance characteristics (i.e., accuracy, speed, computation) of the object detector influence the rate of technology implementation and adoption in agriculture. Thus, the study aims to collect extensive literature to document and critically evaluate the advances and application of YOLO for agricultural object recognition. First, we conducted a bibliometric review of 257 articles to understand the scholarly landscape of YOLO in agricultural domain. Secondly, we conducted a systematic review of 30 articles to identify current knowledge, gaps, and modifications in YOLO for specific agricultural tasks. The study critically assesses and summarizes the information on YOLO's end-to-end learning approach, including data acquisition, processing, network modification, integration, and deployment. We also discussed task-specific YOLO algorithm modification and integration to meet the agricultural object or environment-specific challenges. In general, YOLO-integrated digital tools and technologies show the potential for real-time, automated monitoring, surveillance, and object handling to reduce labor, production cost, and environmental impact while maximizing resource efficiency. The study provides detailed documentation and significantly advances the existing knowledge on applying YOLO in agriculture, which can greatly benefit the scientific community.
Daniel Lindenschmitt, Christoph Fischer, Simon Haussmann
et al.
The transforming process in the scope of agriculture towards Smart Agriculture is an essential step to fulfill growing demands in respect to nourishment. Crucial challenges include establishing robust wireless communication in rural areas, enabling collaboration among agricultural machines, and integrating artificial intelligence into farming practices. Addressing these challenges necessitates a consistent communication system, with wireless communication emerging as a key enabler. Cellular technologies, as 5G and its successor 6G, can offer a comprehensive solution here. Leveraging technologies following the ITU-R M. 2160 recommendation like THz communication, low-latency wireless AI, and embedded sensing, can provide a flexible and energy-efficient infrastructure. This paper introduces on-demand networks based on the OpenRAN approach and a 7.2 functional split. By implementing THz front-hauling between components, a flexible application of 5G or future 6G networks can be realized. Experiments demonstrate that THz communication is suitable for data transmission over the eCPRI interface, particularly in terms of data rate, thereby reducing the need for wired alternatives such as fiber optic cables. Furthermore, limitations such as limited range are discussed, and possible initial solutions are presented. The integration of the OpenRAN standard further enhances flexibility, which is crucial in dynamic agricultural environments. This research contributes to the ongoing discourse on the transformative potential of 6G-enabled wireless communication in shaping the future of smart agriculture.
O objetivo do artigo é demonstrar as mudanças demográficas que ocorreram no estado do Paraná e nos municípios polo da Mesorregião Oeste Paranaense, Cascavel, Foz do Iguaçu e Toledo, por meio das medidas de fecundidade (Taxa de Fecundidade Total (TFT), Taxa Líquida de Reprodução (TLR)), de mortalidade (tabela de vida e expectativa de vida ao nascer), de envelhecimento populacional e estrutura etária (Índice de Envelhecimento (IE) e pirâmides etárias), nos anos de 2000, 2010 e 2022. Os resultados corroboram as teorias e indicam que existe uma relação entre os fatores socioeconômicos e as medidas demográficas encontradas. Foz do Iguaçu apresenta maior fecundidade, as menores expectativas de vida ao nascer e uma população mais jovem, bem como indicadores de desenvolvimento mais defasados. O contrário ocorre com Toledo, que apresenta menor fecundidade, maior expectativa de vida ao nascer, uma população mais envelhecida e melhores indicadores de desenvolvimento. O município de Cascavel e o estado do Paraná apresentam uma situação intermediária.
Abstract: The objective of the paper is to demonstrate the demographic changes that occurred in the state of Paraná and in the pole municipalities of the West Parana Mesoregion, Cascavel, Foz do Iguaçu and Toledo, through fertility measures (Total Fertility Rate (TFR), Net Reproduction Rate Reproduction (NRR)), mortality (life table and life expectancy at birth), population aging and age structure (Aging Index (IE) and age pyramids), in the years 2000, 2010 and 2022. The results corroborate theories and indicate that there is a relationship between socioeconomic factors and the demographic measures found. Foz do Iguaçu has higher fertility, lower life expectancy at birth and a younger population, as well as lagging development indicators. The opposite occurs in Toledo, which has lower fertility, higher life expectancy at birth, an older population and better development indicators. The municipality of Cascavel and the state of Paraná present an intermediate situation.
Rural households live on income much lower than the national average and experience income inequality much higher than the general population. This excess inequality is primarily due to the internal heterogeneity caused by the different nature of household income sources. The purpose of the study was then to assess the level of rural household income inequality and to decompose the inequality index by the main sources of income. The chosen inequality index was Theil-T. The research drew on unidentifiable microdata from the Household Budget Survey conducted by the CSO in 2019-2021.The study found that rural household inequality was slightly higher than that of all Polish households over the analyzed period. Among the various income-source groups, the highest inequality affected farmer households. This group also contributed most to the overall level of inequality in rural areas (44% in 2019 and over 46% in 2021). The pandemic saw an increase in inequality for all identified groups of rural households (the largest – for farmer households) and a decrease in between-group inequality.
Md. Abdul Awal, Pronab Kumar Paul Partha, Md Rafiul Islam
Rodent is one of the major stored pests in Bangladesh, which causes a huge problem in the food sector. They substantially harm to crops and other valuable items on a large scale. Chemically rodent repellents already exist, but because of their toxicity and high price, they are not recommended for use in food storage. The aim of the study was to develop an electronic rodent-repelling circuit that is capable of generating variable ultrasonic frequencies. These frequencies cause discomfort for pests including rats, and nocturnal insects, which affects their aural senses. A device was successfully designed and constructed by microcontroller, printed circuit board, passive infrared sensor, liquid crystal display, Relay module, and other relative components. The device was vigorously tested in Precision Lab, Bangladesh Agricultural University. Laboratory tests were conducted to evaluate the device's repellent efficacy, considering factors such as range, coverage area, power consumption, and durability. The device functioned as it detected the presence of rodents and then released frequency by the positive signal for 30 s. The device generated ultrasonic signals within the range of 20–125 kHz and detected signals from any live substances up to 4.5 m with an error of a maximum of ±0.428 %. The frequency range of rodent hearing is approximately 20 Hz to 300 kHz, with the greatest sensitivity between 30 and 70 kHz, affecting their aural senses and causing discomfort. Therefore, the developed ultrasonic rodent-repellent circuit can be a practical and sustainable solution for repelling stored rodents, which mitigates storage losses caused by rodent infestations.
Adopting drought-tolerant (DT) cultivars is an effective strategy to sustain maize (Zea mays L.) production under water shortage. Optimizing plant density is an important management practice for improving maize yield. In a two-year field trial, the response of yield, actual evapotranspiration (ETc act), and water productivity (WP) to plant density (6, 7.5, 9 plants m−2) was assessed under irrigated and rainfed conditions using a DT (ZD958) and a drought-susceptible (DS, ZY309) maize cultivar, and additionally, the comparison of soil water depletion will be conducted among soils growing different DT maize varieties. Under rainfed, average yield, ETc act, and WP were 24.7%, 8.6% and 14.8% greater in ZD958 than ZY309, respectively. When density increased from 6 to 9 plants m−2, for ZD958 and ZY309 ETc act remained relatively constant, whereas their yield and WP first increased and then decreased and ultimately reached their maximum at 7.5 plants m−2. Under irrigation, increasing density (6–9 plants m−2) significantly increased yield and WP for ZD958, but for ZY309, yield and WP were not significantly impacted. Yield across seasons did not differ between cultivars at 6 and 7.5 plants m−2, and ZD958 had a 10.2% yield advantage over ZY309 at 9 plants m−2. The findings imply that DT cultivar showed greater high density tolerance than DS cultivar and thus higher optimal density under irrigation. Under rainfed, both cultivars had similar density tolerance and optimum density, whereas DT cultivar had stronger drought tolerance than DS cultivar, which could explain DT cultivar’s greater yield and WP. This study indicate that DT cultivar showed higher and more stable yields than DS cultivar across rainfed and irrigated conditions when grown at optimal densities. Thus, sustainable maize production could be achieved by adopting DT cultivars and optimizing density for different conditions in the study region.
Arezoo Jafari, Priscila De Azevedo Drummond, Dominic Nishigaya
et al.
Agricultural workers are essential to the supply chain for our daily food and yet, many face harmful work conditions, including garnished wages, and other labor violations. Workers on H-2A visas are particularly vulnerable due to the precarity of their immigration status being tied to their employer. Although worksite inspections are one mechanism to detect such violations, many labor violations affecting agricultural workers go undetected due to limited inspection resources. In this study, we identify multiple state and industry level factors that correlate with H-2A violations identified by the U.S. Department of Labor Wage and Hour Division using a multilevel zero-inflated negative binomial model. We find that three state-level factors (average farm acreage size, the number of agricultural establishments with less than 20 employees, and higher poverty rates) are correlated with H-2A violations. These findings provide guidance for inspection agencies regarding how to prioritize their limited resources to more effectively inspect agricultural workplaces, thereby improving workplace conditions for H-2A workers.
Mohammad Raziuddin Chowdhury, Md Sakib Ullah Sourav, Rejwan Bin Sulaiman
From the perspective of any nation, rural areas generally present a comparable set of problems, such as a lack of proper health care, education, living conditions, wages, and market opportunities. Some nations have created and developed the concept of smart villages during the previous few decades, which effectively addresses these issues. The landscape of traditional agriculture has been radically altered by digital agriculture, which has also had a positive economic impact on farmers and those who live in rural regions by ensuring an increase in agricultural production. We explored current issues in rural areas, and the consequences of smart village applications, and then illustrate our concept of smart village from recent examples of how emerging digital agriculture trends contribute to improving agricultural production in this chapter.
In this study we argue that integrating ChatGPT into the data processing pipeline of automated sensors in precision agriculture has the potential to bring several benefits and enhance various aspects of modern farming practices. Policy makers often face a barrier when they need to get informed about the situation in vast agricultural fields to reach to decisions. They depend on the close collaboration between agricultural experts in the field, data analysts, and technology providers to create interdisciplinary teams that cannot always be secured on demand or establish effective communication across these diverse domains to respond in real-time. In this work we argue that the speech recognition input modality of ChatGPT provides a more intuitive and natural way for policy makers to interact with the database of the server of an agricultural data processing system to which a large, dispersed network of automated insect traps and sensors probes reports. The large language models map the speech input to text, allowing the user to form its own version of unconstrained verbal query, raising the barrier of having to learn and adapt oneself to a specific data analytics software. The output of the language model can interact through Python code and Pandas with the entire database, visualize the results and use speech synthesis to engage the user in an iterative and refining discussion related to the data. We show three ways of how ChatGPT can interact with the database of the remote server to which a dispersed network of different modalities (optical counters, vibration recordings, pictures, and video), report. We examine the potential and the validity of the response of ChatGPT in analyzing, and interpreting agricultural data, providing real time insights and recommendations to stakeholders