Search Results (1,269)

Residues in the grain tank of seed-production wheat harvesters often cause varietal admixture, challenging seed purity maintenance above 99%. To address this, an intelligent cleaning system was developed for automatic residue recognition and removal. The system utilizes an RGB-D camera and an embedded AI unit paired with an improved lightweight object detection model. This model, enhanced for feature extraction and compressed via LAMP, was successfully deployed on a Jetson Nano, achieving 92.5% detection accuracy and 13.37 FPS for real-time 3D localization of impurities. A D–H kinematic model was established for the 4-DOF cleaning manipulator. By integrating the PSO and FWA models, the motion trajectory was optimized for time-optimality, reducing movement time from 9 s to 5.96 s. Furthermore, a gas–solid coupled simulation verified the separation capability of the cyclone-type dust extraction unit, which prevents motor damage and centralizes residue collection. Field tests confirmed the system’s comprehensive functionality, achieving an average cleaning rate of 92.6%. The proposed system successfully enables autonomous residue cleanup, effectively minimizing the risk of variety mixing and significantly improving the harvest purity and operational reliability of seed-production wheat. It presents a novel technological path for efficient seed production under the paradigm of smart agriculture. Full article

Digital twins (DTs) are increasingly used in controlled-environment agriculture to model microclimates and drive energy-efficient control. However, long-term drift and seasonal variability require continuous recalibration and controller retuning. We develop a self-adaptive DT of a phytotron chamber that combines an MAPE-K loop with an evolutionary heuristic. A genetic algorithm (GA) calibrates the DT parameters against IoT time series and subsequently optimizes heater control settings (two-position, three-position, and proportional modes) subject to comfort constraints on temperature and humidity. Six monitoring intervals (May–June 2025) are used for per-interval calibration and six-fold cross-validation. The calibrated DT reproduces temperature and humidity with high fidelity across unseen intervals: the average cross-validated deviations are 0.27 °C and 7.1%RH (30 transfers). Controller optimization yields cumulative energy savings of 186.54 kWh (3.24%) over six intervals, with per-interval savings ranging from 0.37% to 5.94%. Coupling GA-based DT calibration with model-in-the-loop controller optimization consistently reduces energy use while maintaining microclimate quality, providing a practical pathway for the robust, year-round operation of vertical farms. Full article

This study presents an in-depth exploration of an innovative monitoring system that contributes to precision agriculture (PA) and supports sustainability and biodiversity. Amidst the challenges of global population growth and the need for sustainable, high-yield agricultural practices, PA, supported by modern technology and data-driven methodologies, emerges as a pivotal approach for optimizing crop yield and resource management. The proposed monitoring system integrates Wireless sensor networks (WSNs) into PA, enabling real-time acquisition of environmental data and multimodal observations through cameras and microphones, with data transmission via LTE and/or LoRaWAN for cloud-based analysis. Its main contribution is a physically modular, pole-mounted station architecture that simplifies sensor integration and reconfiguration across use cases, while remaining solar-powered for long-term off-grid operation. The system was evaluated in two field deployments, including a year-long wild-flora monitoring campaign (three stations; 365 days; 1870 images; 63–100% image-based operational availability), during which stations remained operational through a wildfire event. In the viticulture deployment, the acoustic module supported bat monitoring as a bio-indicator of ecosystem health, achieving bat call detection performance of 0.94 (AP Det) and species classification performance of 0.85 (mAP Class). Overall, the results support the use of modular, energy-aware monitoring stations to perform sustained agricultural and ecological data collection under practical field constraints. Full article

Agricultural mechanization and advanced technology have developed significantly in the coffee industry. However, there are still requirements for human laborers to operate, monitor crop health care, and manage production. The integration of advanced technology can significantly enhance the production efficiency and management practices of agricultural enterprises. This paper aims to address these gaps by proposing and implementing a computer vision-based attendance tracking system on mobile platforms that are suitable for the requirements and limitations of agricultural enterprises. First, the face detection process involves interpreting and locating facial structure. Next, the model transforms a photographic image of a human face into digital data based on the unique features and facial structure. We utilize the InsightFace model with the buffalo_l variant, as well as ArcFace with a ResNet backbone, as a facial recognition algorithm. After capturing a facial image, the system conducts a matching process against the existing database to verify identity. Finally, we implement a mobile application prototype on both iOS and Android platforms, ensuring accessibility for farm workers. As a result, our system achieved 95.2% accuracy on the query set, with an average processing time of <200 ms per image (including face detection, embedding extraction, and database matching). The system performs real-time attendance monitoring, automatically recording the entry and exit times of farm workers using facial recognition technology, and enables quick registration of new workers. Our work is expected to enhance transparency and fairness in the human management process, focusing on the coffee farm use case. Full article

The automation of quality control in agricultural food processing, particularly the detection of incomplete peeling in taro, constitutes a critical frontier for ensuring food safety and optimizing production efficiency in the Industry 4.0 era. However, this domain is fraught with significant technical challenges, primarily stemming from the inherent visual characteristics of residual peel: extremely minute scales relative to the vegetable body, highly irregular morphological variations, and the dense occlusion of objects on industrial conveyor belts. To address these persistent impediments, this study introduces a comprehensive solution comprising a specialized dataset and a novel detection architecture. We established the Taro Peel Industrial Dataset (TPID), a rigorously annotated collection of 18,341 high-density instances reflecting real-world production conditions. Building upon this foundation, we propose MDB-YOLO, a lightweight, multi-dimensional bionic detection model evolved from the YOLOv8s architecture. The MDB-YOLO framework integrates a synergistic set of innovations designed to resolve specific detection bottlenecks. To mitigate the conflict between background texture interference and tiny target detection, we integrated the C2f_EMA module with a Wise-IoU (WIoU) loss function, a combination that significantly enhances feature response to low-contrast residues while reducing the penalty on low-quality anchor boxes through a dynamic non-monotonic focusing mechanism. To effectively manage irregular peel shapes, a dynamic feature processing chain was constructed utilizing DySample for morphology-aware upsampling, BiFPN_Concat2 for weighted multi-scale fusion, and ODConv2d for geometric preservation. Furthermore, to address the issue of missed detections caused by dense occlusion in industrial stacking scenarios, Soft-NMS was implemented to replace traditional greedy suppression mechanisms. Experimental validation demonstrates the superiority of the proposed framework. MDB-YOLO achieves a mean Average Precision (mAP50-95) of 69.7% and a Recall of 88.0%, significantly outperforming the baseline YOLOv8s and advanced transformer-based models like RT-DETR-L. Crucially, the model maintains high operational efficiency, achieving an inference speed of 1.1 ms on an NVIDIA A100 and reaching 27 FPS on an NVIDIA Jetson Xavier NX using INT8 quantization. These findings confirm that MDB-YOLO provides a robust, high-precision, and cost-effective solution for real-time quality control in agricultural food processing, marking a significant advancement in the application of computer vision to complex biological targets. Full article

To contribute to the United Nations’ 17 Sustainable Development Goals (SDGs), this study focuses on improving two specific goals—SDG 2 (Zero Hunger) and SDG 12 (Responsible Consumption and Production)—by examining how off-farm employment affects agricultural green total factor productivity (GTFP) in China, a key link between rural socio-economic transformation and agricultural sustainability. The results show that: First, the part-time operation of farmers significantly reduces the green total factor productivity, and the negative impact is more pronounced for off-farm employment households with higher non-agricultural income shares. It mainly stems from the redundant input of land and machinery elements. Second, the effect showed obvious heterogeneous effects at different stages of family development and land management scale. In addition, the scale effect of continuous agricultural production services and the technological synergy effect driven by the deepening of agricultural division of labor are the key to improving green total factor productivity and alleviating the negative effects of part-time operations. In summary, promoting sustainable agricultural practices requires the government to further deepen the reform of the land property rights system and optimize the agricultural socialization service system to ensure both food security and environmental sustainability. Full article

The complexity of field environments poses significant challenges for the segmentation of cotton top leaves, a critical step for apical bud localization in intelligent topping systems. Conventional segmentation models typically rely on large annotated datasets and high computational costs to achieve high precision and robustness. To address these challenges, this paper proposes an efficient and accurate segmentation model, CF-SAM, built upon the Segment Anything Model (SAM) framework. CF-SAM integrates a lightweight Tiny-ViT encoder to reduce computational overhead and employs a LoRA-based fine-tuning strategy for domain adaptation, achieving improved performance with minimal parameter increments. In addition, an Adaptive Prompting Strategy (APS) is introduced to automatically generate high-quality point prompts, enabling fully automated and end-to-end instance segmentation. Trained on only 1000 field images, CF-SAM achieves 98.0% mask accuracy and an mAP@0.5 of 97.83%, while maintaining real-time inference at 58 FPS with only 0.091 M (0.8%) additional parameters. These results demonstrate that CF-SAM achieves an excellent balance between segmentation accuracy and computational cost, providing a reliable technical foundation for apical bud localization and precision agricultural operations. Full article

Autonomous Unmanned Aerial Vehicles (UAVs) are widely used in smart agriculture, logistics, and warehouse management, where precise trajectory prediction is important for safety and efficiency. Traditional approaches require complex physical modeling including mass properties, moment of inertia measurements, and aerodynamic coefficient calculations, which creates significant barriers for custom-built UAVs. Existing trajectory prediction methods are primarily designed for motion forecasting from dense historical observations, which are unsuitable for scenarios lacking historical data (e.g., takeoff phases) or requiring trajectory generation from sparse waypoint specifications (4–6 constraint points). This distinction necessitates architectural designs optimized for spatial interpolation rather than temporal extrapolation. To address these limitations, we present a segmented LSTM framework for complete autonomous flight trajectory prediction. Given target waypoints, our architecture decomposes flight operations, predicts different maneuver types, and outputs the complete trajectory, demonstrating new possibilities for mission-level trajectory planning in autonomous UAV systems. The system consists of a global duration predictor (0.124 MB) and five segment-specific trajectory generators (∼1.17 MB each), with a total size of 5.98 MB and can be deployed in various edge devices. Validated on real Crazyflie 2.1 data, our framework demonstrates high accuracy and provides reliable arrival time predictions, with an Average Displacement Error ranging from 0.0252 m to 0.1136 m. This data-driven approach avoids complex parameter configuration requirements, supports lightweight deployment in edge computing environments, and provides an effective solution for multi-UAV coordination and mission planning applications. Full article

Contemporary nurseries of fruit trees and ornamental plants constitute a key component in the production of high-quality planting material. At present, conventional technology dominates in nurseries in Poland and throughout the European Union. It is based on universal agricultural tractors working with numerous specialized machines—typically underutilized—including sprayers, inter-row cultivation equipment, fertilizer spreaders, and tree lifters. This concept entails several limitations and high investment costs. Because of the considerable size and turning radius of such machinery, a dense network of service roads (every 15–18 m) and wide headlands must be maintained. These areas, which constitute approximately 20% of the total surface, are effectively wasted yet require continuous agronomic maintenance. An alternative concept employs a set of implements mounted on a high-clearance portal tractor (1.6–1.8 m), forming a specialized unit capable of moving above the rows of nursery crops. The study objective of the research was to evaluate the air distribution generated by an air-jet system installed on a crop-spray boom mounted on a portal sprayer, and to assess spray deposition during treatments in nursery trees. Such a configuration enables the mechanization of a broader range of nursery operations than currently possible, while reducing investment costs compared with conventional technology. One still underutilized technology consists of sprayers with an auxiliary airflow (AA) generated by air sleeves. Mean air velocity was measured in three vertical planes, and they showed lower air velocity between 1.0 m and 5.5 m. Spray deposition on apple nursery trees was assessed using a fluorescent tracer. The experimental design consists of a comparative field experiment with and without air flow support, spraying at two standard working rates (200 and 400 L·ha⁻¹) and determining the application of the liquid to plants in the nursery. The results demonstrated a positive effect of the AA system on deposition. At a travel speed of 6.0 km·h⁻¹ and an application rate of 200 L·ha⁻¹, deposition on the upper leaf surface was 68% higher with the fan engaged. For a 400 L·ha⁻¹ rate, deposition increased by 47%, with both differences statistically significant. The study showed that the nursery sprayer mounted on a high-clearance portal tractor and equipped with an AA system achieved an increase of 58% in spray deposition on the upper leaf surface when the fan was operating at 200 L·ha⁻¹ and 28% at 400 L·ha⁻¹. Substantial differences were found between deposition on the upper and lower leaf surfaces, with the former being 20–30 times greater. Given the complexity of nursery production technology, sprayers that ensure the highest possible biological efficacy and high quality of nursery material will play a pivotal role in its development. At the current stage, AA technology fulfils these requirements. Full article

To address the issue of low fertilizer proportioning accuracy in irrigation and fertilization systems due to neglecting the influence of target ions in raw water, this study designed a high-precision online automatic water–fertilizer mixing device that can directly mix raw water (without water purification treatment) with fertilizer stock solution. This device is capable of preparing mixed fertilizer solutions containing N, K, and Ca elements. It employs ion-selective electrodes and flow meters for online detection and feedback of target ion concentrations in the fertilizer solution and flow rate information, and adopts an online fertilizer mixing control strategy that uses a constant raw water flow rate and a fuzzy PID control method to dynamically adjust the pulse frequency of metering pumps, thereby changing the injection volume of nutrient solution. Simulation and experimental analyses show that the piping system of the device is reasonably designed, ensuring stable and smooth fertilizer injection. The temperature-compensated concentration detection models for the three target ions in the fertilizer solution, constructed using a stepwise fitting method, achieve average relative detection errors of 1.94%, 1.18%, and 2.87% for K⁺, NO₃⁻, and Ca²⁺, respectively. When preparing single-element or mixed fertilizer solutions, the device achieves an average steady-state error of no more than 4% and an average steady-state time of approximately 40 s. Compared with deionized water, the average relative errors for potassium ions, nitrate ions, and calcium ions when preparing fertilizer solutions with raw water are 1.33%, 1.12%, and 1.19%, respectively. Compared with the theoretical errors of fertilizer preparation with raw water, the fertilizer proportioning errors of this device for potassium ions, nitrate ions, and calcium ions can be reduced by a maximum of 10.55%, 66.84%, and 62.71%, respectively, which is superior to the performance requirements for water–fertilizer integration equipment specified in the national industry standard DG/T 274-2024. Additionally, the device achieves accurate and stable fertilizer proportioning with safe and reliable operation during 6 h of continuous operation. This device significantly reduces the impact of raw water on fertilizer proportioning accuracy, improves the adaptability of the device to irrigation water sources, and provides theoretical basis and technical support for water-fertilizer integration systems in cost-sensitive agriculture. Full article

Conventional methods of nano-urea application in maize cultivation, such as tractor-operated gun sprayers, involve high water usage, labor intensity, and operator health risks due to chemical exposure. The drone spraying system ensures precise and automated application of nano-urea with minimal resource use, labor requirement, and operator intervention. However, the efficacy of the drone spraying system for nano-urea application was not evaluated and compared with traditional spraying systems in field conditions. There is a need to evaluate whether drone-based spraying systems can provide an equally effective and more resource-efficient alternative to conventional spraying techniques. Therefore, this study evaluated the agronomic efficacy of a drone-based spraying platform in comparison to conventional tractor-operated gun sprayers for the foliar spray application of nano-urea in the maize crop. Field experiments were conducted during the 2024 Kharif season to evaluate changes in SPAD, NDVI values, and grain yield due to two spray application methods. Both spraying methods showed statistically similar NDVI and SPAD values eight days after nano-urea application, indicating comparable effectiveness in nutrient delivery. Maize yield was also observed to be statistically indistinguishable between the two methods (t (8) = 0.025503, p = 0.9803), with 2912 ± 375 kg/ha (mean ± SE) for the gun sprayer and 2928 ± 503 kg/ha for the drone sprayer treatments. However, the drone system demonstrated significant operational advantages, including 95% water savings and decreased operational time. These findings support the use of drone spraying as a sustainable, safe, and scalable alternative to traditional fertilization application practices in precision agriculture. Full article

The demand for real-time land cover mapping from high-resolution remote sensing (HR-RS) imagery motivates lightweight segmentation models running directly on satellites. By processing on-board and transmitting only fine-grained semantic products instead of massive raw imagery, these models provide timely support for disaster response, environmental monitoring, and precision agriculture. Many recent methods combine convolutional neural networks (CNNs) with Transformers to balance local and global feature modeling, with convolutions as explicit information aggregation modules. Such heterogeneous hybrids may be unnecessary for lightweight models if similar aggregation can be achieved homogeneously, and operator inconsistency complicates optimization and hinders deployment on resource-constrained satellites. Meanwhile, lightweight Transformer components in these architectures often adopt aggressive channel compression and shallow contextual interaction to meet compute budgets, impairing boundary delineation and recognition of small or rare classes. To address this, we propose SatViT-Seg, a lightweight semantic segmentation model with a pure Vision Transformer (ViT) backbone. Unlike CNN-Transformer hybrids, SatViT-Seg adopts a homogeneous two-module design: a Local-Global Aggregation and Distribution (LGAD) module that uses window self-attention for local modeling and dynamically pooled global tokens with linear attention for long-range interaction, and a Bi-dimensional Attentive Feed-Forward Network (FFN) that enhances representation learning by modulating channel and spatial attention. This unified design overcomes common lightweight ViT issues such as channel compression and weak spatial correlation modeling. SatViT-Seg is implemented and evaluated in LuoJiaNET and PyTorch; comparative experiments with existing methods are run in PyTorch with unified training and data preprocessing for fairness, while the LuoJiaNET implementation highlights deployment-oriented efficiency on a graph-compiled runtime. Compared with the strongest baseline, SatViT-Seg improves mIoU by up to 1.81% while maintaining the lowest FLOPs among all methods. These results indicate that homogeneous Transformers offer strong potential for resource-constrained, on-board real-time land cover mapping in satellite missions. Full article

In the conventional agricultural production process, the harvesting of mature fruits is frequently dependent on the observation and labor of workers, a process that is often time-consuming and labor-intensive. This study proposes an enhanced YOLOv7 detection and recognition model that incorporates a cross-spatial-channel 3D attention mechanism, a prediction head, and a weighted bidirectional feature pyramid neck optimization. The motivation for this study is to address the issues of uneven target distribution, mutual occlusion of fruits, and uneven light distribution that are prevalent in harvesting operations within orchards. The experimental findings demonstrate that the proposed model achieves an mAP@0.5–0.95 of 89.3%, representing an enhancement of 8.9% in comparison to the initial network. This method has resolved the issue of detecting and positioning the harvesting manipulator in complex orchard scenarios, thereby providing technical support for unmanned agricultural operations. Full article

While artisanal gold mining (AGM) has been credited as a sector that sustains many households in Zimbabwe, it has at the same time been criticized as the chief driver of ecological degradation and social vulnerability. This study qualitatively examines the environmental and socioeconomic impacts of AGM by conducting in-depth interviews with miners, residents, and policymakers across six central mining districts. The study findings indicate that the use of mercury has resulted in severe contamination of water bodies, while clearing land to pave the way for mining has led to severe deforestation, loss of biodiversity, and declining agricultural productivity due to the loss of fertile soils. It was also found that most AGMs were unregulated, and their unregulated operations have intensified health risks, social inequality, and land-use conflicts with the local community. This study provides an insight into how dependence on AGM has perpetuated a cycle of ecological degradation and poverty among many Zimbabweans. The study, therefore, attempts to combine community narratives with policy analysis, thereby proposing a framework for sustainable AGM in Zimbabwe. This involves advocating for the use of environmentally friendly technologies and promoting participatory environmental governance among all key stakeholders. The study contributes to achieving a balance between economic benefits and environmental management by advancing the discourse on sustainable development and community resilience in resource-dependent economies. Full article

Leaf area index (LAI) is a pivotal biophysical parameter linking vegetation physiological processes and macro-ecological functions. Accurate large-scale LAI estimation is indispensable for agricultural management, climate change research, and ecosystem modeling. However, existing methods fail to efficiently extract integrated spatial-spectral-temporal features and lack targeted modeling of spatio-temporal dependencies, compromising the accuracy of LAI products. To address this gap, we propose STC-DeepLAINet, a Transformer-GCN hybrid deep learning architecture integrating spatio-temporal correlations via the following three synergistic modules: (1) a 3D convolutional neural networks (CNNs)-based spectral-spatial embedding module capturing intrinsic correlations between multi-spectral bands and local spatial features; (2) a spatio-temporal correlation-aware module that models temporal dynamics (by “time periods”) and spatial heterogeneity (by “spatial slices”) simultaneously; (3) a spatio-temporal pattern memory attention module that retrieves historically similar spatio-temporal patterns via an attention-based mechanism to improve inversion accuracy. Experimental results demonstrate that STC-DeepLAINet outperforms eight state-of-the-art methods (including traditional machine learning and deep learning networks) in a 500 m resolution LAI inversion task over China. Validated against ground-based measurements, it achieves a coefficient of determination (R²) of 0.827 and a root mean square error (RMSE) of 0.718, outperforming the GLASS LAI product. Furthermore, STC-DeepLAINet effectively captures LAI variability across typical vegetation types (e.g., forests and croplands). This work establishes an operational solution for generating large-scale high-precision LAI products, which can provide reliable data support for agricultural yield estimation and ecosystem carbon cycle simulation, while offering a new methodological reference for spatio-temporal correlation modeling in remote sensing inversion. Full article