Search Results (478)

The acquisition of plant phenotypic traits is essential for selecting superior varieties, improving crop yield, and supporting precision agriculture and agricultural decision-making. Therefore, it plays a significant role in modern agriculture and plant science research. Traditional manual measurements of phenotypic traits are labor-intensive and inefficient. In contrast, combining 3D reconstruction technologies with autonomous vehicles enables more intuitive and efficient trait acquisition. This study proposes a 3D semantic reconstruction system based on an improved ORB-SLAM3 framework, which is mounted on an unmanned vehicle to acquire phenotypic traits in tomato cultivation scenarios. The vehicle is also equipped with the A * algorithm for autonomous navigation. To enhance the semantic representation of the point cloud map, we integrate the BiSeNetV2 network into the ORB-SLAM3 system as a semantic segmentation module. Furthermore, a two-stage filtering strategy is employed to remove outliers and improve the map accuracy, and OctoMap is adopted to store the point cloud data, significantly reducing the memory consumption. A spherical fitting method is applied to estimate the number of tomato fruits. The experimental results demonstrate that BiSeNetV2 achieves a mean intersection over union (mIoU) of 95.37% and a frame rate of 61.98 FPS on the tomato dataset, enabling real-time segmentation. The use of OctoMap reduces the memory consumption by an average of 96.70%. The relative errors when predicting the plant height, canopy width, and volume are 3.86%, 14.34%, and 27.14%, respectively, while the errors concerning the fruit count and fruit volume are 14.36% and 14.25%. Localization experiments on a field dataset show that the proposed system achieves a mean absolute trajectory error (mATE) of 0.16 m and a root mean square error (RMSE) of 0.21 m, indicating high localization accuracy. Therefore, the proposed system can accurately acquire the phenotypic traits of tomato plants, providing data support for precision agriculture and agricultural decision-making. Full article

To address the challenges of low recognition accuracy and substantial model complexity in crop disease identification models operating in complex field environments, this study proposed a novel hybrid model named ConvTransNet-S, which integrates Convolutional Neural Networks (CNNs) and transformers for crop disease identification tasks. Unlike existing hybrid approaches, ConvTransNet-S uniquely introduces three key innovations: First, a Local Perception Unit (LPU) and Lightweight Multi-Head Self-Attention (LMHSA) modules were introduced to synergistically enhance the extraction of fine-grained plant disease details and model global dependency relationships, respectively. Second, an Inverted Residual Feed-Forward Network (IRFFN) was employed to optimize the feature propagation path, thereby enhancing the model’s robustness against interferences such as lighting variations and leaf occlusions. This novel combination of a LPU, LMHSA, and an IRFFN achieves a dynamic equilibrium between local texture perception and global context modeling—effectively resolving the trade-offs inherent in standalone CNNs or transformers. Finally, through a phased architecture design, efficient fusion of multi-scale disease features is achieved, which enhances feature discriminability while reducing model complexity. The experimental results indicated that ConvTransNet-S achieved a recognition accuracy of 98.85% on the PlantVillage public dataset. This model operates with only 25.14 million parameters, a computational load of 3.762 GFLOPs, and an inference time of 7.56 ms. Testing on a self-built in-field complex scene dataset comprising 10,441 images revealed that ConvTransNet-S achieved an accuracy of 88.53%, which represents improvements of 14.22%, 2.75%, and 0.34% over EfficientNetV2, Vision Transformer, and Swin Transformer, respectively. Furthermore, the ConvTransNet-S model achieved up to 14.22% higher disease recognition accuracy under complex background conditions while reducing the parameter count by 46.8%. This confirms that its unique multi-scale feature mechanism can effectively distinguish disease from background features, providing a novel technical approach for disease diagnosis in complex agricultural scenarios and demonstrating significant application value for intelligent agricultural management. Full article

This study identifies and classifies resistance gene analogues (RGAs) in the genomes of Brassica nigra, Sinapis arvensis and Sinapis alba using the RGAugury pipeline. RGAs were categorised into four main classes: receptor-like kinases (RLKs), receptor-like proteins (RLPs), nucleotide-binding leucine-rich repeat (NLR) proteins and transmembrane-coiled-coil (TM-CC) genes. A total of 4499 candidate RGAs were detected, with species-specific proportions. RLKs were the most abundant across all genomes, followed by TM-CCs and RLPs. The sub-classification of RLKs and RLPs identified LRR-RLKs, LRR-RLPs, LysM-RLKs, and LysM-RLPs. Atypical NLRs were more frequent than typical ones in all species. Atypical NLRs were more frequent than typical ones in all species. We explored the relationship between chromosome size and RGA count using regression analysis. In B. nigra and S. arvensis, larger chromosomes generally harboured more RGAs, while S. alba displayed the opposite trend. Exceptions were observed in all species, where some larger chromosomes contained fewer RGAs in B. nigra and S. arvensis, or more RGAs in S. alba. The distribution and density of RGAs across chromosomes were examined. RGA distribution was skewed towards chromosomal ends, with patterns differing across RGA types. Sequence hierarchical pairwise similarity analysis revealed distinct gene clusters, suggesting evolutionary relationships. The study also identified homologous genes among RGAs and non-RGAs in each species, providing insights into disease resistance mechanisms. Finally, RLKs and RLPs were co-localised with reported disease resistance loci in Brassica, indicating significant associations. Phylogenetic analysis of cloned RGAs and QTL-mapped RLKs and RLPs identified distinct clusters, enhancing our understanding of their evolutionary trajectories. These findings provide a comprehensive view of RGA diversity and genomics in these Brassicaceae species, providing valuable insights for future research in plant disease resistance and crop improvement. Full article

To address the issues of low efficiency and large errors in traditional manual cucumber fruit phenotyping methods, this paper proposes the application of keypoint detection technology for cucumber phenotyping and designs an improved lightweight model called YOLO11m-SCFPose. Based on YOLO11m-pose, the original backbone network is replaced with the lightweight StarNet-S1 backbone, reducing model complexity. Additionally, an improved C3K2_PartialConv neck module is used to enhance information interaction and fusion among multi-scale features while maintaining computational efficiency. The Focaler-IoU loss function is employed to improve keypoint localization accuracy. Results show that the improved model achieves an mAP50-95 of 0.924, with a floating-point operation count (GFLOPs) of 32.1, and reduces the model size to 1.229 × 10⁷ parameters. This model demonstrates better computational efficiency and lower resource consumption, providing an effective lightweight solution for crop phenotypic analysis. Full article

This study investigates arthropod assemblage in cowpea, hemp, and watermelon grown both as monocrops and intercrops using three sampling techniques: direct visual counts, sticky cards, and pan traps. A total of 31,774 arthropods were collected, spanning two classes [Arachnida (0.07%) and Insecta (99.93%)], 11 orders, and 82 families representing diverse functional groups. Arachnids were represented by a single family (Araneae). Among insects, the composition included Diptera (36.81%), Thysanoptera (24.64%), Hemiptera (19.43%), Hymenoptera (11.58%), Coleoptera (6.84%), Lepidoptera (0.076%) and Blattodea, Odonata, Orthoptera, Psocodea (≤0.005%). Roughly 10% of the total arthropods were pollinators, while the remainder were primarily herbivores and predators. Apidae were abundant in all treatments except for watermelon monocrops. Intercropping supported more pollinators, particularly Apidae, Halictidae, and Sarcophagidae. However, herbivores dominated (>50%) in each system, largely due to high presence of thrips and cicadellids. Predators accounted for approximately 30%, with dolichopodids (Diptera) being the most dominant. Watermelon yield increased by 30–60% in the intercrop systems. While intercropping increases overall arthropod abundance, it also creates a more balanced community where beneficial organisms are not heavily outnumbered by pests and contributes to enhanced ecological resilience and crop performance. Full article

A field experiment in growing spring wheat (Triticum aestivum L.—cv. ‘Monsun’) under organic, integrated and conventional farming systems was conducted over the period of 2020–2022 at the Czesławice Experimental Farm (Lubelskie Voivodeship, Poland). The first experimental factor analyzed was the farming system: A. organic system (control)—without the use of chemical plant protection products and NPK mineral fertilization; B. conventional system—the use of plant protection products and NPK fertilization in the range and doses recommended for spring wheat; C. integrated system—use of plant protection products and NPK fertilization in an “economical” way—doses reduced by 50%. The second experimental factor was irrigation strategy: 1. no irrigation—control; 2. double irrigation; 3. multiple irrigation The aim of the research was to determine the physical, chemical, and enzymatic properties of loess soil under spring wheat crops as influenced by the factors listed above. The highest organic C content of the soil (1.11%) was determined in the integrated system with multiple irrigation of spring wheat, whereas the lowest one (0.77%)—in the conventional system without irrigation. In the conventional system, the highest contents of total N (0.15%), P (131.4 mg kg⁻¹), and K (269.6 mg kg⁻¹) in the soil were determined under conditions of multiple irrigation. In turn, the organic system facilitated the highest contents of Mg, B, Cu, Mn, and Zn in the soil, especially upon multiple irrigation of crops. It also had the most beneficial effect on the evaluated physical parameters of the soil. In each farming system, the multiple irrigation of spring wheat significantly increased moisture content, density, and compaction of the soil and also improved its total sorption capacity (particularly in the integrated system). The highest count of beneficial fungi, the lowest population number of pathogenic fungi, and the highest count of actinobacteria were recorded in the soil from the organic system. Activity of soil enzymes was the highest in the integrated system, followed by the organic system—particularly upon multiple irrigation of crops. Summing up, the present study results demonstrate varied effects of the farming systems on the quality and health of loess soil. From a scientific point of view, the integrated farming system ensures the most stable and balanced physicochemical and biological parameters of the soil due to the sufficient amount of nutrients supplied to the soil and the minimized impact of chemical plant protection products on the soil. The multiple irrigation of crops resulting from indications of soil moisture sensors mounted on plots (indicating the real need for irrigation) contributed to the improvement of almost all analyzed soil quality indices. Multiple irrigation generated high costs, but in combination with fertilization and chemical crop protection (conventional and integrated system), it influenced the high productivity of spring wheat and compensated for the incurred costs (the greatest profit). Full article

The flowering stage is a critical phase in the growth of rapeseed crops, and non-destructive, high-throughput quantitative analysis of rape flower clusters in field environments holds significant importance for rapeseed breeding. However, detecting and counting rape flower clusters remains challenging in complex field conditions due to their small size, severe overlapping and occlusion, and the large parameter sizes of existing models. To address these challenges, this study proposes a lightweight rape flower clusters detection model, SPL-YOLOv8. First, the model introduces StarNet as a lightweight backbone network for efficient feature extraction, significantly reducing computational complexity and parameter counts. Second, a feature fusion module (C2f-Star) is integrated into the backbone to enhance the feature representation capability of the neck through expanded spatial dimensions, mitigating the impact of occluded regions on detection performance. Additionally, a lightweight Partial Group Convolution Detection Head (PGCD) is proposed, which employs Partial Convolution combined with Group Normalization to enable multi-scale feature interaction. By incorporating additional learnable parameters, the PGCD enhances the detection and localization of small targets. Finally, channel pruning based on the Layer-Adaptive Magnitude-based Pruning (LAMP) score is applied to reduce model parameters and runtime memory. Experimental results on the Rapeseed Flower-Raceme Benchmark (RFRB) demonstrate that the SPL-YOLOv8n-prune model achieves a detection accuracy of 92.2% in Average Precision (AP₅₀), comparable to SOTA methods, while reducing the giga floating point operations per second (GFLOPs) and parameters by 86.4% and 95.4%, respectively. The model size is only 0.5 MB and the real-time frame rate is 171 fps. The proposed model effectively detects rape flower clusters with minimal computational overhead, offering technical support for yield prediction and elite cultivar selection in rapeseed breeding. Full article

To support the growing global population, sustainable farming methods like vertical farming must complement traditional agriculture. This study evaluated the effects of various nitrogen fertilizer application rates (N_low (1055.3 ppm), N_rec (1640.9 ppm), N_high (2811.3 ppm), and N_0 (469.9 ppm)) on organic kale (Brassica oleracea L. var. acephala ‘Lacinato’) and Swiss chard (Beta vulgaris subsp. Vulgaris ‘Ruby/Rhubarb Red’), grown in a vertical growing system installed in a high tunnel during the spring and fall season of 2023 at the organic farm of Tennessee State University. Growth parameters studied included fresh weight, Brix, chlorophyll, plant height, and leaf count. Most parameters did not exhibit statistically significant differences (alpha = 0.05). However, consistent numerical trends and deviations were observed. Although not statistically significant, kale achieved the highest mean fresh weight in N_rec (688.08 g), and Swiss chard in N_high by spring (649.62 g). Among the few parameters, significant differences were observed for Swiss chard plant height (48.07 cm) and leaf count (47.25), with N_high during fall. Findings suggest that while definitive conclusions were limited, recommended nitrogen rates (N_rec) may enhance crop performance and contribute sustainable yields in resource constrained vertical farming systems. Further controlled studies are warranted to validate trends and refine nutrient strategies in vertical growing system. Full article

This study focuses on enhancing crop yield prediction in plant factory environments through precise crop canopy image capture and background interference removal. This method achieves highly accurate recognition of the crop canopy projection area (CCPA), with a coefficient of determination (R²) of 0.98. A spatial resolution of 0.078 mm/pixel was derived by referencing a scale ruler and processing pixel counts, eliminating outliers in the data. Image post-processing focused on extracting the canopy boundary and calculating the crop canopy area. By incorporating crop yield data, a comparative analysis of 28 prediction models was performed, assessing performance metrics such as MSE, RMSE, MAE, MAPE, R², prediction speed, training time, and model size. Among them, the Wide Neural Network model emerged as the most optimal. It demonstrated remarkable predictive accuracy with an R² of 0.95, RMSE of 27.15 g, and MAPE of 11.74%. Furthermore, the model achieved a high prediction speed of 60,234.9 observations per second, and its compact size of 7039 bytes makes it suitable for efficient, real-time deployment in practical applications. This model offers substantial support for managing crop growth, providing a solid foundation for refining cultivation processes and enhancing crop yields. Full article

Banana is an important cash and food crop worldwide. Recent outbreaks of banana diseases are threatening the global banana industry and smallholder livelihoods. Remote sensing data offer the potential to detect the presence of disease, but formal analysis is needed to compare inferred disease data with observed disease data. In this study, we present a novel remote-sensing-based framework that combines Landsat-8 imagery with meteorology-informed phenological models and machine learning to identify anomalies in banana crop health. Unlike prior studies, our approach integrates domain-specific crop phenology to enhance the specificity of anomaly detection. We used a pixel-level random forest (RF) model to predict 11 key vegetation indices (VIs) as a function of historical meteorological conditions, specifically daytime and nighttime temperature from MODIS and precipitation from NASA GES DISC. By training on periods of healthy crop growth, the RF model establishes expected VI values under disease-free conditions. Disease presence is then detected by quantifying the deviations between observed VIs from Landsat-8 imagery and these predicted healthy VI values. The model demonstrated robust predictive reliability in accounting for seasonal variations, with forecasting errors for all VIs remaining within 10% when applied to a disease-free control plantation. Applied to two documented outbreak cases, the results show strong spatial alignment between flagged anomalies and historical reports of banana bunchy top disease (BBTD) and Fusarium wilt Tropical Race 4 (TR4). Specifically, for BBTD in Australia, a strong correlation of 0.73 was observed between infection counts and the discrepancy between predicted and observed NDVI values at the pixel with the highest number of infections. Notably, VI declines preceded reported infection rises by approximately two months. For TR4 in Mozambique, the approach successfully tracked disease progression, revealing clear spatial spread patterns and correlations as high as 0.98 between VI anomalies and disease cases in some pixels. These findings support the potential of our method as a scalable early warning system for banana disease detection. Full article

Antimicrobial use in food animals selects for antimicrobial-resistant (AMR) bacteria, which most commonly reach humans via the food chain. However, AMR bacteria can also escape the feedyard via agricultural runoff, manure used as crop fertilizer, and even dust. A study published in 2015 reported AMR genes in dust from cattle feedyards; however, one of the study’s major limitations was the failure to investigate gene presence in viable bacteria, or more importantly, viable bacteria of importance to human health. Our main objective was to investigate the presence and quantity of viable bacteria and antimicrobial-resistant (AMR) determinants in fugitive bioaerosols from cattle feedyards in the downwind environment. Six bioaerosol sampling campaigns were conducted at three commercial beef cattle feedyards to assess variability in viable bacteria and AMR determinants associated with geographic location, meteorological conditions, and season. Dust samples were collected using four different sampling methods, and spiral plated in triplicate on both non-selective and antibiotic-selective media. Colonies of total aerobic bacteria, Enterococcus spp., Salmonella enterica, and Escherichia coli were enumerated. Viable bacteria, including AMR bacteria, were identified in dust from cattle feedyards. Bacteria and antimicrobial resistance genes (ARGs via qPCR) were mainly found in downwind samples. Total suspended particles (TSPs) and impinger samples yielded the highest bacterial counts. Genes encoding beta-lactam resistance (bla_CMY-2 and bla_CTX-M) were detected while the most common ARG was tet(M). The predominant Salmonella serovar identified was Lubbock. Further research is needed to assess how far viable AMR bacteria can travel in the ambient environment downwind from cattle feedyards, to model potential public health risks. Full article

Wireworms are a major threat to potatoes. Agronomic prevention is always the first IPM strategy to be implemented. This work assesses whether a bioactive cover crop, sunn hemp (Crotalaria juncea L.), a tropical leguminous plant, reduces wireworm damage risk when cultivated as a crop preceding potatoes. The effects of Crotalaria plants (alive, chopped, and incorporated) on wireworms and tuber-damage prevention were studied in semi-natural (pots) and open-field conditions. The survival of a set number of reared wireworms feeding on Crotalaria plants or potato tubers in soil with incorporated Crotalaria chopped tissues was assessed. Wireworm damage on tubers was assessed in fields where Crotalaria had been cultivated, chopped, and incorporated the previous year. The tuber damage assessment involved counting all the erosions/scars caused by wireworm feeding. The prevalent wireworm species studied was Agriotes sordidus. Our research is the first to demonstrate that Crotalaria as a cover crop can significantly reduce potato damage by wireworms. A major role is likely played by the high pyrrolizidine alkaloid content in Crotalaria juncea tissues, but this has to be specifically proven. Crotalaria juncea may thus represent an effective means for use alone or with complementary ones to produce potatoes with low wireworm damage without using synthetic insecticides. Full article

Object detection is essential for precision agriculture applications like automated plant counting, but the minimum dataset requirements for effective model deployment remain poorly understood for arable crop seedling detection on orthomosaics. This study investigated how much annotated data is required to achieve standard counting accuracy (R² = 0.85) for maize seedlings across different object detection approaches. We systematically evaluated traditional deep learning models requiring many training examples (YOLOv5, YOLOv8, YOLO11, RT-DETR), newer approaches requiring few examples (CD-ViTO), and methods requiring zero labeled examples (OWLv2) using drone-captured orthomosaic RGB imagery. We also implemented a handcrafted computer graphics algorithm as baseline. Models were tested with varying training sources (in-domain vs. out-of-distribution data), training dataset sizes (10–150 images), and annotation quality levels (10–100%). Our results demonstrate that no model trained on out-of-distribution data achieved acceptable performance, regardless of dataset size. In contrast, models trained on in-domain data reached the benchmark with as few as 60–130 annotated images, depending on architecture. Transformer-based models (RT-DETR) required significantly fewer samples (60) than CNN-based models (110–130), though they showed different tolerances to annotation quality reduction. Models maintained acceptable performance with only 65–90% of original annotation quality. Despite recent advances, neither few-shot nor zero-shot approaches met minimum performance requirements for precision agriculture deployment. These findings provide practical guidance for developing maize seedling detection systems, demonstrating that successful deployment requires in-domain training data, with minimum dataset requirements varying by model architecture. Full article

The increasing global demand for resilient, sustainable agricultural systems has intensified the need for advanced monitoring strategies, particularly for climate-adaptive crops such as Moringa oleifera Lam. This study presents an integrated approach using Unmanned Aerial Vehicles (UAVs) equipped with multispectral and thermal cameras to monitor the vegetative performance and determine the optimal harvest period of four M. oleifera genotypes in a Mediterranean environment. High-resolution data were collected and processed to generate the NDVI, canopy temperature, and height maps, enabling the assessment of plant vigor, stress conditions, and spatial canopy structure. NDVI analysis revealed robust vegetative growth (0.7–0.9), with optimal harvest timing identified on 30 October 2024, when the mean NDVI exceeded 0.85. Thermal imaging effectively discriminated plant crowns from surrounding weeds by capturing cooler canopy zones due to active transpiration. A clear inverse correlation between NDVI and Land Surface Temperature (LST) was observed, reinforcing its relevance for stress diagnostics and environmental monitoring. The results underscore the value of UAV-based multi-sensor systems for precision agriculture, offering scalable tools for phenotyping, harvest optimization, and sustainable management of medicinal and aromatic crops in semiarid regions. Moreover, in this study, to produce M. oleifera leaf powder intended for use as a food ingredient, the leaves of four M. oleifera genotypes were dried, milled, and evaluated for their hygiene and safety characteristics. Plate count analyses confirmed the absence of pathogenic bacterial colonies in the M. oleifera leaf powders, highlighting their potential application as natural and functional additives in food production. Full article

Background: Soybean is an important global crop in food security and agricultural economics. Accurate estimation of soybean pod counts is critical for yield prediction, breeding programs, precision farming, etc. Traditional methods, such as manual counting, are slow, labor-intensive, and prone to errors. With rapid advancements in artificial intelligence (AI), deep learning has enabled automatic pod number estimation in collaboration with unmanned aerial vehicles (UAVs). However, existing AI models are computationally demanding and require significant processing resources (e.g., memory). These resources are often not available in rural regions and small farms. Methods: To address these challenges, this study presents a set of lightweight, efficient AI models designed to overcome these limitations. By integrating model simplification, weight quantization, and squeeze-and-excitation (SE) self-attention blocks, we develop compact AI models capable of fast and accurate soybean pod count estimation. Results and Conclusions: Experimental results show a comparable estimation accuracy of 84–87%, while the AI model size is significantly reduced by a factor of 9–65, thus making them suitable for deployment in edge devices, such as Raspberry Pi. Compared to existing models such as YOLO POD and SoybeanNet, which rely on over 20 million parameters to achieve approximately 84% accuracy, our proposed lightweight models deliver a comparable or even higher accuracy (84.0–86.76%) while using fewer than 2 million parameters. In future work, we plan to expand the dataset by incorporating diverse soybean images to enhance model generalizability. Additionally, we aim to explore more advanced attention mechanisms—such as CBAM or ECA—to further improve feature extraction and model performance. Finally, we aim to implement the complete system in edge devices and conduct real-world testing in soybean fields. Full article