Hasil untuk "Agriculture (General)"

J. Yienger, H. Levy

E. Silbergeld, Jay P. Graham, L. Price

Fabrice Ofridam, Mohamad Tarhini, N. Lebaz et al.

Stimuli-responsive materials in general and pH-responsive polymers in particular have gained increasing interest during the last two decades. Their unique properties, which arise from their ability to exhibit sharp and reversible changes in response to environmental pH conditions, have made them suitable for various applications such as drug delivery and specific body-site targeting, sensing and actuation, membrane functionalization, separation techniques, as well as in agriculture and food industry and even chemical industries. In the present review, the focus is on the general characteristics of pH-responsive polymers in terms of their origin, chemical composition, and preparation. Moreover, some of the important and recent applications are reported and discussed.

S. U, V. Nagaveni, B. Raghavendra

In India, Agriculture plays an essential role because of the rapid growth of population and increased in demand for food. Therefore, it needs to increase in crop yield. One major effect on low crop yield is disease caused by bacteria, virus and fungus. It can be prevented by using plant diseases detection techniques. Machine learning methods can be used for diseases identification because it mainly apply on data themselves and gives priority to outcomes of certain task. This paper presents the stages of general plant diseases detection system and comparative study on machine learning classification techniques for plant disease detection. In this survey it observed that Convolutional Neural Network gives high accuracy and detects more number of diseases of multiple crops.

Ebasa Temesgen, Nathnael Minyelshowa, Lebsework Negash

The use of unmanned aerial vehicles (UAVs) in precision agriculture has seen a huge increase recently. As such, systems that aim to apply various algorithms on the field need a structured framework of abstractions. This paper defines the various tasks of the UAVs in precision agriculture and model them into an architectural framework. The presented architecture is built on the context that there will be minimal physical intervention to do the tasks defined with multiple coordinated and cooperative UAVs. Various tasks such as image processing, path planning, communication, data acquisition, and field mapping are employed in the architecture to provide an efficient system. Besides, different limitation for applying Multi-UAVs in precision agriculture has been considered in designing the architecture. The architecture provides an autonomous end-to-end solution, starting from mission planning, data acquisition and image processing framework that is highly efficient and can enable farmers to comprehensively deploy UAVs onto their lands. Simulation and field tests shows that the architecture offers a number of advantages that include fault-tolerance, robustness, developer and user-friendliness.

E. Nicholls, A. Ely, L. Birkin et al.

Food production depends upon the adequate provision of underpinning ecosystem services, such as pollination. Paradoxically, conventional farming practices are undermining these services and resulting in degraded soils, polluted waters, greenhouse gas emissions and massive loss of biodiversity including declines in pollinators. In essence, farming is undermining the ecosystem services it relies upon. Finding alternative more sustainable ways to meet growing food demands which simultaneously support biodiversity is one of the biggest challenges facing humanity. Here, we review the potential of urban and peri-urban agriculture to contribute to sustainable food production, using the 17 sustainable development goals set by the United Nations General Assembly as a framework. We present new data from a case study of urban gardens and allotments in the city of Brighton and Hove, UK. Such urban and peri-urban landholdings tend to be small and labour-intensive, characterised by a high diversity of crops including perennials and annuals. Our data demonstrate that this type of agricultural system can be highly productive and that it has environmental and social advantages over industrial agriculture in that crops are usually produced using few synthetic inputs and are destined for local consumption. Overall, we conclude that food grown on small-scale areas in and near cities is making a significant contribution to feeding the world and that this type of agriculture is likely to be relatively favourable for some ecosystem services, such as supporting healthy soils. However, major knowledge gaps remain, for example with regard to productivity, economic and employment impacts, pesticide use and the implications for biodiversity.

Mohey Eldeen H. H. Ali, Ahmed F. Tayel, Hossam M. Ezzat et al.

Energy plays a crucial role in national development, influencing critical sectors such as industry, agriculture, healthcare, and education. Accurate energy consumption prediction is essential for efficient energy management, helping prevent imbalances between supply and demand and potential energy shortages. This study aims to forecast the total primary energy supply (TPES), using Egypt as a case study for the first time in literature and utilizing several models (ordinary differential equations (ODEs), regression, and ANN models). Although ordinary differential equations (ODEs) offer flexibility and convenience, their application in energy forecasting remains limited. One of the main objectives of this research is to evaluate the effectiveness of ODEs in predicting energy consumption. Various ODE and regression models are employed to identify the most suitable model amongst each category for forecasting energy demand. Additionally, an artificial neural network (ANN) is developed, trained, validated, and tested for the same forecasting task. The study compares the performance of the selected ODE model (Mendelsohn), with the selected regression model (Polynomial), and an ANN model predicting Egypt’s TPES until 2035. By assessing multiple forecasting methods, this work improves the accuracy and reliability of energy consumption predictions, which is crucial for sustainable energy planning and policy development.

Xiangdong Xu, Lin Chen, Hewei Meng et al.

To explore the vibration transmission characteristics of jujube mechanical harvesting, and optimize the relationship between vibration input and dynamic response of jujube branches, the vibration characteristics simulation and layered vibration test of jujube branches were carried out. The jujube branch model was established by means of three-dimensional scanning and reverse reconstruction. The natural frequency and suitable vibration parameter range of the jujube branch model were obtained by simulation. Finally, the stratified vibration field experiment of jujube branch was carried out. The results show that there are multi-order natural frequencies of jujube branch in the range of 0–30 Hz. The typical vibration modes include the overall deformation of jujube branch, the deformation of unilateral branch and the deformation of the end of twigs. The resonance frequencies of the measuring points on different branches are mostly close, but the frequencies of the maximum peaks on different paths are different, which is often related to the branch path. The optimal working parameter combination under layered vibration is: the lower layer excitation frequency and amplitude are 5.80 Hz and 7.00 mm, the upper layer excitation frequency and amplitude are 15.60 Hz and 8.50 mm. Under this parameter combination, the acceleration of the measuring point on the fine branch is closest to the separation acceleration. Under this parameter combination, the average harvest rate is 88.74 %. The research can provide reference for the development of forest fruit vibration harvesting machinery.

Patrick Michael, Robert J. Reid, Robert W. Fitzpatrick

The long-term roles of live plant roots in mitigating acid sulfate soil stresses remain poorly understood. Three studies, each lasting twelve months, were conducted using Melaleuca armillaris and Phragmites australis. In the first study, alkaline sandy loam soil was mixed into the sulfuric soil to increase the pH to 6.7, and Melaleuca seedlings were planted. In the second and third studies, M. armillaris and P. australis were planted in sulfuric and sulfidic soils and maintained at 75% water-holding capacity and flooded soil conditions. All the studies were set using 300 mm stormwater tubes with sealed bottom ends. The treatments were replicated four times, set up under a glasshouse in a completely randomized design, and harvested after 12 months. The pH and root biomass were measured from the surface, middle, and deep profiles. Results showed that the neutralization obtained by mixing alkaline sandy loam soil with sulfuric soil was stable but deteriorated due to plant root penetration. In the sulfuric soil material (pH <4), M. armillaris produced more roots at the surface than in the deep soil under circumneutral pH and aerobic soil conditions. In sulfidic soil material (pH >4), more roots were produced in the deeper soils. In the sulfuric and sulfidic soil materials, P. australis produced more roots at the surface than at the deep under pH >4 and aerobic conditions. Under anaerobic conditions with a pH >4, root distribution was even. Our findings suggest that common terrestrial and aquatic plants maintain a characteristic distribution of roots to mitigate the stresses of acid sulfate soils.

Mihir Gupta, Abhay Mangla, Ross Greer et al.

Precision agriculture relies heavily on accurate image analysis for crop disease identification and treatment recommendation, yet existing vision-language models (VLMs) often underperform in specialized agricultural domains. This work presents a domain-aware framework for agricultural image processing that combines prompt-based expert evaluation with self-consistency mechanisms to enhance VLM reliability in precision agriculture applications. We introduce two key innovations: (1) a prompt-based evaluation protocol that configures a language model as an expert plant pathologist for scalable assessment of image analysis outputs, and (2) a cosine-consistency self-voting mechanism that generates multiple candidate responses from agricultural images and selects the most semantically coherent diagnosis using domain-adapted embeddings. Applied to maize leaf disease identification from field images using a fine-tuned PaliGemma model, our approach improves diagnostic accuracy from 82.2\% to 87.8\%, symptom analysis from 38.9\% to 52.2\%, and treatment recommendation from 27.8\% to 43.3\% compared to standard greedy decoding. The system remains compact enough for deployment on mobile devices, supporting real-time agricultural decision-making in resource-constrained environments. These results demonstrate significant potential for AI-driven precision agriculture tools that can operate reliably in diverse field conditions.

Chee Mei Ling, Thangarajah Akilan, Aparna Ravinda Phalke

Agricultural image semantic segmentation is a pivotal component of modern agriculture, facilitating accurate visual data analysis to improve crop management, optimize resource utilization, and boost overall productivity. This study proposes an efficient image segmentation method for precision agriculture, focusing on accurately delineating farmland anomalies to support informed decision-making and proactive interventions. A novel Dual Atrous Separable Convolution (DAS Conv) module is integrated within the DeepLabV3-based segmentation framework. The DAS Conv module is meticulously designed to achieve an optimal balance between dilation rates and padding size, thereby enhancing model performance without compromising efficiency. The study also incorporates a strategic skip connection from an optimal stage in the encoder to the decoder to bolster the model's capacity to capture fine-grained spatial features. Despite its lower computational complexity, the proposed model outperforms its baseline and achieves performance comparable to highly complex transformer-based state-of-the-art (SOTA) models on the Agriculture Vision benchmark dataset. It achieves more than 66% improvement in efficiency when considering the trade-off between model complexity and performance, compared to the SOTA model. This study highlights an efficient and effective solution for improving semantic segmentation in remote sensing applications, offering a computationally lightweight model capable of high-quality performance in agricultural imagery.

George Grispos, Logan Mears, Larry Loucks et al.

As technology increasingly integrates into farm settings, the food and agriculture sector has become vulnerable to cyberattacks. However, previous research has indicated that many farmers and food producers lack the cybersecurity education they require to identify and mitigate the growing number of threats and risks impacting the industry. This paper presents an ongoing research effort describing a cybersecurity initiative to educate various populations in the farming and agriculture community. The initiative proposes the development and delivery of a ten-module cybersecurity course, to create a more secure workforce, focusing on individuals who, in the past, have received minimal exposure to cybersecurity education initiatives.

Brian Gopalan, Nathalia Nascimento, Vishal Monga

This paper addresses the critical need for efficient and accurate weed segmentation from drone video in precision agriculture. A quality-aware modular deep-learning framework is proposed that addresses common image degradation by analyzing quality conditions-such as blur and noise-and routing inputs through specialized pre-processing and transformer models optimized for each degradation type. The system first analyzes drone images for noise and blur using Mean Absolute Deviation and the Laplacian. Data is then dynamically routed to one of three vision transformer models: a baseline for clean images, a modified transformer with Fisher Vector encoding for noise reduction, or another with an unrolled Lucy-Richardson decoder to correct blur. This novel routing strategy allows the system to outperform existing CNN-based methods in both segmentation quality and computational efficiency, demonstrating a significant advancement in deep-learning applications for agriculture.

Risa Shinoda, Nakamasa Inoue, Hirokatsu Kataoka et al.

Precise automated understanding of agricultural tasks such as disease identification is essential for sustainable crop production. Recent advances in vision-language models (VLMs) are expected to further expand the range of agricultural tasks by facilitating human-model interaction through easy, text-based communication. Here, we introduce AgroBench (Agronomist AI Benchmark), a benchmark for evaluating VLM models across seven agricultural topics, covering key areas in agricultural engineering and relevant to real-world farming. Unlike recent agricultural VLM benchmarks, AgroBench is annotated by expert agronomists. Our AgroBench covers a state-of-the-art range of categories, including 203 crop categories and 682 disease categories, to thoroughly evaluate VLM capabilities. In our evaluation on AgroBench, we reveal that VLMs have room for improvement in fine-grained identification tasks. Notably, in weed identification, most open-source VLMs perform close to random. With our wide range of topics and expert-annotated categories, we analyze the types of errors made by VLMs and suggest potential pathways for future VLM development. Our dataset and code are available at https://dahlian00.github.io/AgroBenchPage/ .

Madhav Rijal, Rashik Shrestha, Trevor Smith et al.

This study presents a methodology to safely manipulate branches to aid various agricultural tasks. Humans in a real agricultural environment often manipulate branches to perform agricultural tasks effectively, but current agricultural robots lack this capability. This proposed strategy to manipulate branches can aid in different precision agriculture tasks, such as fruit picking in dense foliage, pollinating flowers under occlusion, and moving overhanging vines and branches for navigation. The proposed method modifies RRT* to plan a path that satisfies the branch geometric constraints and obeys branch deformable characteristics. Re-planning is done to obtain a path that helps the robot exert force within a desired range so that branches are not damaged during manipulation. Experimentally, this method achieved a success rate of 78% across 50 trials, successfully moving a branch from different starting points to a target region.

Aruna Gauba, Irene Pi, Yunze Man et al.

We present AgMMU, a challenging real-world benchmark for evaluating and advancing vision-language models (VLMs) in the knowledge-intensive domain of agriculture. Unlike prior datasets that rely on crowdsourced prompts, AgMMU is distilled from 116,231 authentic dialogues between everyday growers and USDA-authorized Cooperative Extension experts. Through a three-stage pipeline: automated knowledge extraction, QA generation, and human verification, we construct (i) AgMMU, an evaluation set of 746 multiple-choice questions (MCQs) and 746 open-ended questions (OEQs), and (ii) AgBase, a development corpus of 57,079 multimodal facts covering five high-stakes agricultural topics: insect identification, species identification, disease categorization, symptom description, and management instruction. Benchmarking 12 leading VLMs reveals pronounced gaps in fine-grained perception and factual grounding. Open-sourced models trail after proprietary ones by a wide margin. Simple fine-tuning on AgBase boosts open-sourced model performance on challenging OEQs for up to 11.6% on average, narrowing this gap and also motivating future research to propose better strategies in knowledge extraction and distillation from AgBase. We hope AgMMU stimulates research on domain-specific knowledge integration and trustworthy decision support in agriculture AI development.

Bo Yang, Yunkui Chen, Lanfei Feng et al.

Despite rapid advances in multimodal large language models, agricultural applications remain constrained by the scarcity of domain-tailored models, curated vision-language corpora, and rigorous evaluation. To address these challenges, we present the AgriGPT-VL Suite, a unified multimodal framework for agriculture. Our contributions are threefold. First, we introduce Agri-3M-VL, the largest vision-language corpus for agriculture to our knowledge, curated by a scalable multi-agent data generator; it comprises 1M image-caption pairs, 2M image-grounded VQA pairs, 50K expert-level VQA instances, and 15K GRPO reinforcement learning samples. Second, we develop AgriGPT-VL, an agriculture-specialized vision-language model trained via a progressive curriculum of textual grounding, multimodal shallow/deep alignment, and GRPO refinement. This method achieves strong multimodal reasoning while preserving text-only capability. Third, we establish AgriBench-VL-4K, a compact yet challenging evaluation suite with open-ended and image-grounded questions, paired with multi-metric evaluation and an LLM-as-a-judge framework. Experiments show that AgriGPT-VL outperforms leading general-purpose VLMs on AgriBench-VL-4K, achieving higher pairwise win rates in the LLM-as-a-judge evaluation. Meanwhile, it remains competitive on the text-only AgriBench-13K with no noticeable degradation of language ability. Ablation studies further confirm consistent gains from our alignment and GRPO refinement stages. We will open source all of the resources to support reproducible research and deployment in low-resource agricultural settings.

C. Arndt, Felix Ankomah Asante, J. Thurlow

Long-run economic development in Ghana is potentially vulnerable to anthropogenic climate change given the country’s dependence on rain-fed agriculture, hydropower and unpaved rural roads. We use a computable general equilibrium model, informed by detailed sector studies, to estimate the economy-wide impacts of climate change under four climate projections. Climate change is found to always reduce national welfare, with poor and urban households and the northern Savannah zone being the worst affected. However, there is wide variation across scenarios in the size of climate impacts and in the relative importance of sectoral impact channels, thus underscoring the need for multi-sector approaches that account for climate uncertainty. Our analysis of adaptation options indicates that investing in agricultural research and extension, and improved road surfaces, are potentially cost-effective means of mitigating most of the damages from climate change in Ghana.

Mirelly Caroline Alves, Suellen Barbara Ferreira Galvino-Costa, Priscilla de Sousa Geraldino Duarte et al.

ABSTRACT Potato virus Y (PVY) is recognized as one of the most common and destructive pathogens seriously affecting potato producing areas worldwide. More recently PVYNTN and PVYN:O/N-Wi, have emerged as the main strains present in the PVY infected plants detected in Brazilian potato fields. In this study, samples of potato collected in south part of Minas Gerais - Brazil were first tested by DAS-ELISA and then by RT-PCR multiplex in order to discriminate the PVY strains. Afterward, part of them was tested by RT-qPCR to confirm and quantify the viruses in infected tissues. The sensitivity of the techniques for detecting PVY isolates present in the sampled locations was investigated, as well as the occurrence of mixed infections, aiming to understand the general epidemiological picture of this pathogen in potato producing fields. In the multiplex RT-PCR test, the samples with O and N serotypes were identified as infected with PVYNTN and PVYN:O/N-Wi strains. When tested by RT-PCR for amplification of PVYE, 41 samples (67,2%) were positive, having a characteristic electrophoretic profile for this recombinant strain, and 9 isolates were also observed with atypical patterns for recombinant PVYE. The best technique to detect mixed infection was RT-qPCR, with the concentration of PVYNTN being much higher than that of PVYN:O/N-Wi. These results show the importance of using the most suitable method for the diagnosis and surveying of PVY strains in crop fields and reveal, for the first time, the dissemination of PVYE recombinants to several Brazilian potato fields.

Pratyush Tripathy, Kathy Baylis, Kyle Wu et al.

Accurate mapping of agricultural field boundaries is crucial for enhancing outcomes like precision agriculture, crop monitoring, and yield estimation. However, extracting these boundaries from satellite images is challenging, especially for smallholder farms and data-scarce environments. This study explores the Segment Anything Model (SAM) to delineate agricultural field boundaries in Bihar, India, using 2-meter resolution SkySat imagery without additional training. We evaluate SAM's performance across three model checkpoints, various input sizes, multi-date satellite images, and edge-enhanced imagery. Our results show that SAM correctly identifies about 58% of field boundaries, comparable to other approaches requiring extensive training data. Using different input image sizes improves accuracy, with the most significant improvement observed when using multi-date satellite images. This work establishes proof of concept for using SAM and maximizing its potential in agricultural field boundary mapping. Our work highlights SAM's potential in delineating agriculture field boundary in training-data scarce settings to enable a wide range of agriculture related analysis.