Search Results (8)

This study presents a cost-effective and accurate method for assessing nitrogen concentration in vermicompost fertilizer using a low-cost TCS3200 color sensor and a K-Nearest Neighbors (KNN) machine learning model. The objective was to evaluate the performance of four different color spaces—RGB, Lab, LCh, and CMYK—identify the most effective feature representation for a multi-class classification task based on accuracy and theoretical robustness to ambient light variations. A total of 2400 data points were collected from a standard chemical test kit and processed. A rigorous 60-fold cross-validation approach was used to determine the optimal model hyperparameters and to ensure the robustness of the findings. The results demonstrate that the model trained on the LCh color space achieved the highest classification accuracy of 0.9708 with an optimal K-value of 6, significantly outperforming Lab (0.9688), RGB (0.9625), and CMYK (0.9583). A detailed analysis of the confusion matrix revealed that the model successfully classified the ‘High’ and ‘Medium’ nitrogen levels with near-perfect accuracy, while minor misclassifications occurred between the ‘Low’ and ‘Trace’ categories (5 Low ⟶ Trace, 6 Trace ⟶ Low). The proposed system offers a practical, robust, and accessible tool for precision agriculture, enabling farmers to make informed decisions regarding fertilization, and directly supporting sustainable agriculture and responsible resource management. The findings indicate that the LCh color space is highly effective for this application, providing a viable solution for the rapid and reliable assessment of vermicompost quality. Most importantly, this inexpensive, on-site system removes the need for costly, time-consuming laboratory analyses, giving farmers and compost users the instantaneous, accurate nitrogen data they need to maximize crop yield, optimize nutrient application, and significantly reduce input costs from overfertilization. Full article

In this article, the authors developed a novel method for the moisture mapping of the soil surface of agrophytocenosis using a neural network based on synchronized radar and multispectral optoelectronic data from Sentinel-1,2. The significance of this research lies in its potential to enhance precision farming practices, which are increasingly vital in addressing global agricultural challenges such as water scarcity and the need for sustainable resource management. To verify the developed method, data from two experimental plots were utilized. These plots were located on irrigated soybean crops, with the first plot situated on the right bank (plot No. 1) and the second on the left bank (plot No. 2) of the lower Volga River. Two experimental soil moisture geodatasets were created through measurements and geo-referencing points using the gravimetric method (for plot No. 1) and the proximal sensing method (for plot No. 2) employing the Soil Moisture Sensor ML3-KIT (THETAKIT, Delta). The soil moisture retrieval algorithm was based on the use of a neural network to predict the reflection coefficient of an electro-magnetic wave from the soil surface, followed by inversion into soil moisture using a dielectric model that takes into account the soil texture. The input parameter of the neural network was the ratio of the microwave radar vegetation index (calculated based on Sentinel-1 data) to the index (calculated based on the data of multispectral optoelectronic channels 8 and 11 of Sentinel-2). The retrieved soil moisture values were compared with in situ measurements, showing a determination coefficient of 0.44–0.65 and a standard deviation of 2.4–4.2% for plot No. 1 and similar metrics for plot No. 2. The conducted research laid the groundwork for developing a new technology for remote sensing of soil moisture content in agrophytocenosis, serving as a crucial component of precision farming systems and agroecology. The integration of this technology promotes sustainable agricultural practices by minimizing water consumption while maximizing crop productivity. This aligns with broader environmental goals of conserving natural resources and reducing agricultural runoff. On a larger scale, data derived from such studies can inform policy decisions related to water resource management, guiding regulations that promote efficient water use in agriculture. Full article

Smart agricultural drones for crop spraying are becoming popular worldwide. Research institutions, commercial companies, and government agencies are investigating and promoting the use of technologies in the agricultural industry. This study presents a smart agriculture drone integrated with Internet of Things technologies that use machine learning techniques such as TensorFlow Lite with an EfficientDetLite1 model to identify objects from a custom dataset trained on three crop classes, namely, pineapple, papaya, and cabbage species, achieving an inference time of 91 ms. The system’s operation is characterised by its adaptability, offering two spray modes, with spray modes A and B corresponding to a 100% spray capacity and a 50% spray capacity based on real-time data, embodying the potential of Internet of Things for real-time monitoring and autonomous decision-making. The drone is operated with an X500 development kit and has a payload of 1.5 kg with a flight time of 25 min, travelling at a velocity of 7.5 m/s at a height of 2.5 m. The drone system aims to improve sustainable farming practices by optimising pesticide application and improving crop health monitoring. Full article

Sustainable food systems have the potential to protect humans and planet health. Green public procurement (GPP) is a tool for the sustainable transformation. In Poland, the share of GPP is extremely low. As part of the StratKIT project, a survey-based research study was carried out in the city of Rybnik (Silesia Region). The aim of this paper is to diagnose the level of awareness in the field of sustainable development of the project stakeholders, and to propose further sustainable actions related to GPP in Poland. The survey was conducted in social care homes and two primary schools. Statistical analyses were performed using SPSS 24 software. The results show that the level of education has an impact on the assessment of the environment, and that the place of residency interferes with the level of environmental, organic and nutritional knowledge. Correlational analysis showed no statistically significant relationships between age, level of education, place of residence and willingness to introduce action connected to GPP (e.g., organic food). In conclusion, there is a need for an appropriate educational program for the public procurement and catering services (PPCS) sector, teaching about advantages of GPP for the food systems in connection to sustainable agriculture, consumption and climate actions. Full article

The Internet of Things (IoT) has rapidly progressed in recent years and immensely influenced many industries in how they operate. Consequently, IoT technology has improved productivity in many sectors, and smart farming has also hugely benefited from the IoT. Smart farming enables precision agriculture, high crop yield, and the efficient utilization of natural resources to sustain for a longer time. Smart farming includes sensing capabilities, communication technologies to transmit the collected data from the sensors, and data analytics to extract meaningful information from the collected data. These modules will enable farmers to make intelligent decisions and gain profits. However, incorporating new technologies includes inheriting security and privacy consequences if they are not implemented in a secure manner, and smart farming is not an exception. Therefore, security monitoring is an essential component to be implemented for smart farming. In this paper, we propose a cloud-enabled smart-farm security monitoring framework to monitor device status and sensor anomalies effectively and mitigate security attacks using behavioral patterns. Additionally, a blockchain-based smart-contract application was implemented to securely store security-anomaly information and proactively mitigate similar attacks targeting other farms in the community. We implemented the security-monitoring-framework prototype for smart farms using Arduino Sensor Kit, ESP32, AWS cloud, and the smart contract on the Ethereum Rinkeby Test Network and evaluated network latency to monitor and respond to security events. The performance evaluation of the proposed framework showed that our solution could detect security anomalies within real-time processing time and update the other farm nodes to be aware of the situation. Full article

A national animal disease surveillance network initiated by the Lao PDR government is adopted and reinforced by a joint research project between the National Animal Health Laboratory (NAHL), the Department of Livestock and Fisheries (DLF), and the Mahidol Oxford Tropical Medicine Research Unit (MORU). The network is strengthened by staff training and practical exercises and is utilised to provide zoonotic or high-impact disease information on a national scale. Between January and December 2020, large ruminant samples are collected monthly from 18 abattoirs, one in each province, by provincial and district agriculture and forestry officers. The surveillance network collected a total of 4247 serum samples (1316 buffaloes and 2931 cattle) over this period. Samples are tested for antibodies against Brucella spp., Coxiella burnetii (Q fever) and Foot-and-Mouth Disease Non-Structural Protein (FMD NSP) using commercial ELISA kits and the Rose Bengal test. Seroprevalences of Q fever and brucellosis in large ruminants are low at 1.7% (95% CI: 1.3, 2.1) and 0.7% (95% CI: 0.5, 1.0) respectively, while for FMD NSP it is 50.5% (95% CI: 49.0, 52.0). Univariate analyses show differences in seroprevalences of Q fever between destination (abattoir) province (p-value = 0.005), province of origin (p-value = 0.005), animal type (buffalo or cattle) (p-value = 0.0008), and collection month (p-value = 3.4 × 10⁻⁶). Similar to Q fever, seroprevalences of brucellosis were significantly different for destination province (p-value < 0.00001), province of origin (p-value < 0.00001), animal type (p-value = 9.9 × 10⁻⁵) and collection month (p-value < 0.00001), plus body condition score (p-value = 0.003), and age (p-value = 0.007). Additionally, risk factors of the FMD NSP dataset include the destination province (p-value < 0.00001), province of origin (p-value < 0.00001), sex (p-value = 7.97 × 10⁻⁸), age (p-value = 0.009), collection date (p-value < 0.00001), and collection month (p-value < 0.00001). Spatial analyses revealed that there is no spatial correlation of FMD NSP seropositive animals. High-risk areas for Q fever and brucellosis are identified by spatial analyses. Further investigation of the higher risk areas would provide a better epidemiological understanding of both diseases in Lao PDR. In conclusion, the abattoir serological survey provides useful information about disease exposure and potential risk factors. The network is a good base for field and laboratory staff training in practical technical skills. However, the sustainability of such a surveillance activity is relatively low without an external source of funding, given the operational costs and insufficient government budget. The cost-effectiveness of the abattoir survey could be increased by targeting hotspot areas, reducing fixed costs, and extending the focus to cover more diseases. Full article

Printed pictures are traditional forms of agricultural extension for smallholder farmers. They receive historical academic criticism but remain inexpensive, do not require technical skills (unlike smartphones), and bypass language/literacy barriers. Here, a comprehensive participatory pipeline is described that included 56 Nepalese women farmer editors to develop 100 picture-based lessons. Thereafter, the Theory of Planned Behavior is used as a framework to evaluate 20 diverse lessons using quantitative and qualitative data (Nvivo-11) collected from four groups, focusing on low-literacy women: the women farmer editors (n = 56); smallholder field testers who had prior exposure to extension agents and the actual innovations (control group, n = 120), and those who did not (test group, n = 60); expert stakeholders (extension agents/scientists, n = 25). The expected comprehension difference between farmer groups was non-substantive, suggesting that the participatory editing/testing approaches were effective. There were surprising findings compared to the academic literature: smallholders comprehended the pictures without the help of extension agents, perhaps because of the participatory approaches used; children assisted their mothers to understand caption-based lessons; the farmers preferred printed pictures compared to advanced information and communication technologies (ICTs); and the resource-poor farmers were willing to pay for the printed materials, sufficient to make them cost-neutral/scalable. These findings have implications for smallholder farmers beyond Nepal. Full article

Soil volumetric water content ( $VWC$ ) is a vital parameter to understand several ecohydrological and environmental processes. Its cost-effective measurement can potentially drive various technological tools to promote data-driven sustainable agriculture through supplemental irrigation solutions, the lack of which has contributed to severe agricultural distress, particularly for smallholder farmers. The cost of commercially available $VWC$ sensors varies over four orders of magnitude. A laboratory study characterizing and testing sensors from this wide range of cost categories, which is a prerequisite to explore their applicability for irrigation management, has not been conducted. Within this context, two low-cost capacitive sensors—SMEC300 and SM100—manufactured by Spectrum Technologies Inc. (Aurora, IL, USA), and two very low-cost resistive sensors—the Soil Hygrometer Detection Module Soil Moisture Sensor (YL100) by Electronicfans and the Generic Soil Moisture Sensor Module (YL69) by KitsGuru—were tested for performance in laboratory conditions. Each sensor was calibrated in different repacked soils, and tested to evaluate accuracy, precision and sensitivity to variations in temperature and salinity. The capacitive sensors were additionally tested for their performance in liquids of known dielectric constants, and a comparative analysis of the calibration equations developed in-house and provided by the manufacturer was carried out. The value for money of the sensors is reflected in their precision performance, i.e., the precision performance largely follows sensor costs. The other aspects of sensor performance do not necessarily follow sensor costs. The low-cost capacitive sensors were more accurate than manufacturer specifications, and could match the performance of the secondary standard sensor, after soil specific calibration. SMEC300 is accurate ( $MAE$ , $RMSE$ , and $RAE$ of 2.12%, 2.88% and 0.28 respectively), precise, and performed well considering its price as well as multi-purpose sensing capabilities. The less-expensive SM100 sensor had a better accuracy ( $MAE$ , $RMSE$ , and $RAE$ of 1.67%, 2.36% and 0.21 respectively) but poorer precision than the SMEC300. However, it was established as a robust, field ready, low-cost sensor due to its more consistent performance in soils (particularly the field soil) and superior performance in fluids. Both the capacitive sensors responded reasonably to variations in temperature and salinity conditions. Though the resistive sensors were less accurate and precise compared to the capacitive sensors, they performed well considering their cost category. The YL100 was more accurate ( $MAE$ , $RMSE$ , and $RAE$ of 3.51%, 5.21% and 0.37 respectively) than YL69 ( $MAE$ , $RMSE$ , and $RAE$ of 4.13%, 5.54%, and 0.41, respectively). However, YL69 outperformed YL100 in terms of precision, and response to temperature and salinity variations, to emerge as a more robust resistive sensor. These very low-cost sensors may be used in combination with more accurate sensors to better characterize the spatiotemporal variability of field scale soil moisture. The laboratory characterization conducted in this study is a prerequisite to estimate the effect of low- and very low-cost sensor measurements on the efficiency of soil moisture based irrigation scheduling systems. Full article