Abstract Plant diseases have a disastrous impact on the safety of food production, and they can cause a significant reduction in both the quality and quantity of agricultural products. In severe cases, plant diseases may even cause no grain harvest entirely. Thus, the automatic identification and diagnosis of plant diseases are highly desired in the field of agricultural information. Many methods have been proposed for solving this task, where deep learning is becoming the preferred method due to the impressive performance. In this work, we study transfer learning of the deep convolutional neural networks for the identification of plant leaf diseases and consider using the pre-trained model learned from the typical massive datasets, and then transfer to the specific task trained by our own data. The VGGNet pre-trained on ImageNet and Inception module are selected in our approach. Instead of starting the training from scratch by randomly initializing the weights, we initialize the weights using the pre-trained networks on the large labeled dataset, ImageNet. The proposed approach presents a substantial performance improvement with respect to other state-of-the-art methods; it achieves a validation accuracy of no less than 91.83% on the public dataset. Even under complex background conditions, the average accuracy of the proposed approach reaches 92.00% for the class prediction of rice plant images. Experimental results demonstrate the validity of the proposed approach, and it is accomplished efficiently for plant disease detection.
Significance Overuse of agricultural chemicals has resulted in enormous damages to environmental quality and human health in China. Reducing the use of agricultural chemicals to an optimal level is a crucial challenge for the sustainable development of agriculture. We demonstrate that small farm size (in China, typically ∼0.1 ha for each parcel) is strongly related to overuse of agricultural chemicals. Farm size increases with economic development in many other countries, but this is not observed in China due to national policies. Increasing farm size by removing policy distortions would substantially decrease both the use of agricultural chemicals and their environmental impact, while increasing rural income in China. Understanding the reasons for overuse of agricultural chemicals is critical to the sustainable development of Chinese agriculture. Using a nationally representative rural household survey from China, we found that farm size is a strong factor that affects the use intensity of agricultural chemicals across farms in China. Statistically, a 1% increase in farm size is associated with a 0.3% and 0.5% decrease in fertilizer and pesticide use per hectare (P < 0.001), respectively, and an almost 1% increase in agricultural labor productivity, while it only leads to a statistically insignificant 0.02% decrease in crop yields. The same pattern was also found using other independently collected data sources from China and an international panel analysis of 74 countries from the 1960s to the 2000s. While economic growth has been associated with increasing farm size in many other countries, in China this relationship has been distorted by land and migration policies, leading to the persistence of small farm size in China. Removing these distortions would decrease agricultural chemical use by 30–50% and the environmental impact of those chemicals by 50% while doubling the total income of all farmers including those who move to urban areas. Removing policy distortions is also likely to complement other remedies to the overuse problem, such as easing farmer’s access to modern technologies and knowledge, and improving environmental regulation and enforcement.
Md. Asif Hossain, Nabil Subhan, Mantasha Rahman Mahi
et al.
Access to reliable agricultural advisory remains limited in many developing regions due to a persistent language barrier: authoritative agricultural manuals are predominantly written in English, while farmers primarily communicate in low-resource local languages such as Bengali. Although recent advances in Large Language Models (LLMs) enable natural language interaction, direct generation in low-resource languages often exhibits poor fluency and factual inconsistency, while cloud-based solutions remain cost-prohibitive. This paper presents a cost-efficient, cross-lingual Retrieval-Augmented Generation (RAG) framework for Bengali agricultural advisory that emphasizes factual grounding and practical deployability. The proposed system adopts a translation-centric architecture in which Bengali user queries are translated into English, enriched through domain-specific keyword injection to align colloquial farmer terminology with scientific nomenclature, and answered via dense vector retrieval over a curated corpus of English agricultural manuals (FAO, IRRI). The generated English response is subsequently translated back into Bengali to ensure accessibility. The system is implemented entirely using open-source models and operates on consumer-grade hardware without reliance on paid APIs. Experimental evaluation demonstrates reliable source-grounded responses, robust rejection of out-of-domain queries, and an average end-to-end latency below 20 seconds. The results indicate that cross-lingual retrieval combined with controlled translation offers a practical and scalable solution for agricultural knowledge access in low-resource language settings
Agricultural environments present high proportions of spatially dense navigation bottlenecks for long-term navigation and operational planning of agricultural mobile robots. The existing agent-centric multi-robot path planning (MRPP) approaches resolve conflicts from the perspective of agents, rather than from the resources under contention. Further, the density of such contentions limits the capabilities of spatial interleaving, a concept that many planners rely on to achieve high throughput. In this work, two variants of the priority-based Fragment Planner (FP) are presented as resource-centric MRPP algorithms that leverage route fragmentation to enable partial route progression and limit the impact of binary-based waiting. These approaches are evaluated in lifelong simulation over a 3.6km topological map representing a commercial polytunnel environment. Their performances are contrasted against 5 baseline algorithms with varying robotic fleet sizes. The Fragment Planners achieved significant gains in throughput compared with Prioritised Planning (PP) and Priority-Based Search (PBS) algorithms. They further demonstrated a task throughput of 95% of the optimal task throughput over the same time period. This work shows that, for long-term deployment of agricultural robots in corridor-dominant agricultural environments, resource-centric MRPP approaches are a necessity for high-efficacy operational planning.
Access to credit is essential for transforming South Africa’s agricultural sector, particularly in enhancing value chain sustainability. This study investigated the role of credit access in supporting smallholder farmers’ value chain sustainability, as part of a broader project focused on developing a credit risk model for South African farmers. Data were collected from 223 SAFDA farmers in KwaZulu-Natal and Mpumalanga using a structured questionnaire. The average treatment effects (ATEs) of a propensity score matching (PSM) model was used to estimate the impacts of credit on the following four key variables: farm ownership, farm size, farm income, and farm assets. The results showed that farm ownership was associated with credit access, as ownership provided 1.84 times the chances of loan approval. Additionally, farm income increased by ZAR 2,849,398 for credit recipients compared to non-recipients. This income boost enhances market linkages and food value chain sustainability. This study rejects the hypothesis that credit access has no impact on smallholder farmers, highlighting its vital role in promoting agricultural development and value chain growth. It is recommended that policymakers enhance credit access and risk mitigation strategies to further support smallholder farmers. To improve access to credit for smallholder farmers in South Africa, we recommend the following measures: (1) establishing credit guarantee schemes in partnership with financial institutions to reduce lending risks; and (2) implementing financial education programs for smallholder farmers to enhance their debt management skills. Credit access has the potential to promote positive change across economic, social, and environmental aspects, improving not only the livelihoods of smallholder farmers but also contributing to broader sustainable development goals.
Ancient tea plantations possess extremely important economic and cultivation value. In China, ancient tea plantations with trees over 100 years old have been preserved. However, the status of soil microorganisms, soil fertility, and soil heavy metal pollution in these ancient tea plantations remains unclear. This study took four Dancong ancient tea plantations in Fenghuang, Chaozhou City, and Guangdong Province as the research objects. Soil samples were collected from the surface layer (0–20 cm) and subsurface layer (20–40 cm) of the ancient tea trees. The rhizosphere soil microbial diversity and soil nutrients were determined. On this basis, the soil fertility was evaluated by referring to the soil environmental quality standards so as to conduct a comprehensive evaluation of the soil in the Dancong ancient tea plantations. This study found that <i>Proteobacteria</i>, <i>Acidobacteriota</i>, <i>Chloroflexi</i>, and <i>Actinobacteria</i> were the dominant bacteria in the rhizosphere soil of the Dancong ancient tree tea plantation. <i>Ascomycota</i> and <i>Mortierellomycota</i> are the dominant fungal phyla. <i>Subgroup_2</i>, <i>AD3</i>, <i>Acidothermus</i>, and <i>Acidibacter</i> were the dominant bacterial genera. <i>Saitozyma</i>, <i>Mortierella</i>, and <i>Fusarium</i> are the dominant fungal genera. The redundancy analysis (RDA) revealed that at the bacterial phylum level, <i>Verrucomicrobia</i> showed positive correlations with alkali-hydrolyzable nitrogen (AN), available potassium (AK), and total nitrogen (TN); <i>Proteobacteria</i> exhibited a positive correlation with available phosphorus (AP); and <i>Gemmatimonadetes</i> was positively correlated with total potassium (TK). At the fungal phylum level, <i>Ascomycota</i> demonstrated a positive correlation with TK. TN, AN, and TK were identified as key physicochemical indicators influencing soil bacterial diversity, while TN, AN, AP, and AK were the key physicochemical indicators affecting soil fungal diversity. This study revealed that the soil of Dancong ancient tea plantations has reached Level I fertility in terms of TN, TP, SOM, and AP. TK and AN show Level I or near-Level I fertility, but AK only meets Level III fertility for tea planting, serving as the main limiting factor for soil fertility quality. Considering the relatively abundant TK content in the tea plantations, potassium-solubilizing bacteria should be prioritized over blind potassium fertilizer application. Meanwhile, it is particularly noteworthy that AN and SOM are at extremely high levels. Sustained excess of AN and SOM may lead to over-proliferation of dominant microorganisms, inhibition of other functional microbial communities, and disruption of ecological balance. Therefore, optimizing nutrient input methods during fertilization is recommended.
Persistent financial frictions - including price volatility, constrained credit access, and supply chain inefficiencies - have long hindered productivity and welfare in the global agricultural sector. This paper provides a theoretical and applied analysis of how fiat-collateralized stablecoins, a class of digital currency pegged to a stable asset like the U.S. dollar, can address these long-standing challenges. We develop a farm-level profit maximization model incorporating transaction costs and credit constraints to demonstrate how stablecoins can enhance economic outcomes by (1) reducing the costs and risks of cross-border trade, (2) improving the efficiency and transparency of supply chain finance through smart contracts, and (3) expanding access to credit for smallholder farmers. We analyze key use cases, including parametric insurance and trade finance, while also considering the significant hurdles to adoption, such as regulatory uncertainty and the digital divide. The paper concludes that while not a panacea, stablecoins represent a significant financial technology with the potential to catalyze a paradigm shift in agricultural economics, warranting further empirical investigation and policy support.
Thien Hoang Nguyen, Erik Muller, Michael Rubin
et al.
Traditional soil sampling and analysis methods are labor-intensive, time-consuming, and limited in spatial resolution, making them unsuitable for large-scale precision agriculture. To address these limitations, we present a robotic solution for real-time sampling, analysis and mapping of key soil properties. Our system consists of two main sub-systems: a Sample Acquisition System (SAS) for precise, automated in-field soil sampling; and a Sample Analysis Lab (Lab) for real-time soil property analysis. The system's performance was validated through extensive field trials at a large-scale Australian farm. Experimental results show that the SAS can consistently acquire soil samples with a mass of 50g at a depth of 200mm, while the Lab can process each sample within 10 minutes to accurately measure pH and macronutrients. These results demonstrate the potential of the system to provide farmers with timely, data-driven insights for more efficient and sustainable soil management and fertilizer application.
Aryan Singh Dalal, Sidharth Rai, Rahul Singh
et al.
Application rate errors when using self-propelled agricultural sprayers for agricultural production remain a concern. Among other factors, spray boom instability is one of the major contributors to application errors. Spray booms' width of 38m, combined with 30 kph driving speeds, varying terrain, and machine dynamics when maneuvering complex field boundaries, make controls of these booms very complex. However, there is no quantitative knowledge on the extent of boom movement to systematically develop a solution that might include boom designs and responsive boom control systems. Therefore, this study was conducted to develop an automated computer vision system to quantify the boom movement of various agricultural sprayers. A computer vision system was developed to track a target on the edge of the sprayer boom in real time. YOLO V7, V8, and V11 neural network models were trained to track the boom's movements in field operations to quantify effective displacement in the vertical and transverse directions. An inclinometer sensor was mounted on the boom to capture boom angles and validate the neural network model output. The results showed that the model could detect the target with more than 90 percent accuracy, and distance estimates of the target on the boom were within 0.026 m of the inclinometer sensor data. This system can quantify the boom movement on the current sprayer and potentially on any other sprayer with minor modifications. The data can be used to make design improvements to make sprayer booms more stable and achieve greater application accuracy.
Stephane Ngnepiepaye Wembe, Vincent Rousseau, Johann Laconte
et al.
In this paper, we present a novel method to control a rigidly connected location on the vehicle, such as a point on the implement in case of agricultural tasks. Agricultural robots are transforming modern farming by enabling precise and efficient operations, replacing humans in arduous tasks while reducing the use of chemicals. Traditionnaly, path_following algorithms are designed to guide the vehicle's center along a predefined trajetory. However, since the actual agronomic task is performed by the implement, it is essential to control a specific point on the implement itself rather than vehicle's center. As such, we present in this paper two approaches for achieving the control of an offset point on the robot. The first approach adapts existing control laws, initially inteded for rear axle's midpoint, to manage the desired lateral deviation. The second approach employs backstepping control techniques to create a control law that directly targets the implement. We conduct real-world experiments, highlighting the limitations of traditional approaches for offset points control, and demonstrating the strengths and weaknesses of the proposed methods.
Cropland non-agriculturalization refers to the conversion of arable land into non-agricultural uses such as forests, residential areas, and construction sites. This phenomenon not only directly leads to the loss of cropland resources but also poses systemic threats to food security and agricultural sustainability. Accurate identification of cropland and non-cropland areas is crucial for detecting and addressing this issue. Traditional CNNs employ static convolution layers, while dynamic convolution studies demonstrate that adaptively weighting multiple convolutional kernels through attention mechanisms can enhance accuracy. However, existing dynamic convolution methods relying on Global Average Pooling (GAP) for attention weight allocation suffer from information loss, limiting segmentation precision. This paper proposes Frequency-Aware Dynamic Convolution (FADConv) and a Frequency Attention (FAT) module to address these limitations. Building upon the foundational structure of dynamic convolution, we designed FADConv by integrating 2D Discrete Cosine Transform (2D DCT) to capture frequency domain features and fuse them. FAT module generates high-quality attention weights that replace the traditional GAP method,making the combination between dynamic convolution kernels more reasonable.Experiments on the GID and Hi-CNA datasets demonstrate that FADConv significantly improves segmentation accuracy with minimal computational overhead. For instance, ResNet18 with FADConv achieves 1.9% and 2.7% increases in F1-score and IoU for cropland segmentation on GID, with only 58.87M additional MAdds. Compared to other dynamic convolution approaches, FADConv exhibits superior performance in cropland segmentation tasks.