Bee pollination is pivotal to plant reproduction and crop yield, making its identification and monitoring highly significant for agricultural production. However, practical detection of bee pollination poses various challenges, including the small size of bee targets, their low pixel occupancy in images, and the complexity of floral backgrounds. Aimed to scientifically evaluate pollination efficiency, accurately detect the pollination status of flowers, and provide reliable data to guide flower and fruit thinning in orchards, ultimately supports the scientific management of bee colonies and enhances agricultural efficiency, a lightweight recognition model that can effectively overcome the above obstacles was proposed, thereby advancing the practical application of bee pollination detection technology in smart agriculture.

A specialized bee pollination dataset was constructed comprising three flower types: strawberry, blueberry, and chrysanthemum. High-resolution cameras were used to record videos of the pollination process, which were then subjected to frame sampling to extract representative images. These initial images underwent manual screening to ensure quality and relevance. To address challenges such as limited data diversity and class imbalance, a comprehensive data augmentation strategy was employed. Techniques including rotation, flipping, brightness adjustment, and mosaic augmentation were applied, significantly expanding the dataset's size and variability. The enhanced dataset was subsequently split into training and validation sets at an 8:2 ratio to ensure robust model evaluation. The base detection model was built upon an improved YOLOv10n architecture. The conventional C2f module in the backbone was replaced with a novel cross stage partial network_multi-scale edge information enhance (CSP_MSEE) module, which synergizes the cross-stage partial connections from cross stage partial network (CSPNet) with a multiscale edge enhancement strategy. This design greatly improved feature extraction, particularly in scenarios involving fine-grained structures and small-scale targets like bees. For the neck, a hybrid-scale feature pyramid network (HS-FPN) was implemented, incorporating a channel attention (CA) mechanism and a dimension matching (DM) module to refine and align multi-scale features. These features were further integrated through a selective feature fusion (SFF) module, enabling the effective combination of low-level texture details and high-level semantic representations. The detection head was replaced with the lightweight shared detail enhanced convolutional detection head (LSDECD), an enhanced version of the Lightweight shared convolutional detection head (LSCD) detection head. It incorporated detail enhancement convolution (DEConv) from DEA-Net to improve the extraction of fine-grained bee features.Additionally, the standard convolution_groupnorm (Conv_GN) layers were replaced with detail enhancement convolution_ groupnorm (DEConv_GN), significantly reducing model parameters and enhancing the model's sensitivity to subtle bee behaviors. This lightweight yet accurate model design made it highly suitable for real-time deployment on resource-constrained edge devices in agricultural environments.

Experimental results on the three bee pollination datasets: strawberry, blueberry, and chrysanthemum, demonstrated the effectiveness of the proposed improvements over the baseline YOLOv10n model. The enhanced model achieved significant reductions in computational overhead, lowering the computational complexity by 3.1 GFLOPs and the number of parameters by 1.3 M. The computational cost of the improved model reached 5.1 GFLOPS, and the number of parameters was 1.3 M. These reductions contribute to improved efficiency, making the model more suitable for deployment on edge devices with limited processing capabilities, such as mobile platforms or embedded systems used in agricultural monitoring. In terms of detection performance, the improved model showed consistent gains across all three datasets. Specifically, the recall rates reached 82.6% for strawberry flowers, 84.0% for blueberry flowers, and 84.8% for chrysanthemum flowers. Corresponding mAP50(Mean Average Precision at IoU threshold of 0.5) scores were 89.3%, 89.5%, and 88.0%, respectively. Compared to the original YOLOv10n model, these results marked respective improvements of 2.1% in recall and 1.7% in mAP50 on the strawberry dataset, 2.0% and 2.6% on the blueberry dataset, and 2.1% and 2.2% on the chrysanthemum dataset.

The proposed YOLOv10n-CHL lightweight bee pollination detection model, through coordinated enhancements at multiple architectural levels, achieved notable improvements in both detection accuracy and computational efficiency across multiple bee pollination datasets. The model significantly improved the detection performance for small objects while substantially reducing computational overhead, facilitating its deployment on edge computing platforms such as drones and embedded systems. This research could provide a solid technical foundation for the precise monitoring of bee pollination behavior and the advancement of smart agriculture. Nevertheless, the model's adaptability to extreme lighting and complex weather conditions remains an area for improvement. Future work will focus on enhancing the model's robustness in these scenarios to support its broader application in real-world agricultural environments.

This study aimed to study the effects of different proportions of cow manure and fertilizer on soil labile organic carbon components and carbon conversion related enzyme activities in apple orchard, and to reveal the mechanism of different fertilization methods on biological transformation of soil carbon pool, so as to provide the theoretical support for organic and inorganic scientific application and soil quality improvement in apple orchard.

Long-term positioning fertilization test was used as the platform. Six treatments were selected: no fertilizer (CK), 100% fertilizer (CF100), 25% cow manure with 75% fertilizer (CM25CF75), 50% cow manure with 50% fertilizer (CM50CF50), 75% cow manure with 25% fertilizer (CM75CF25), and 100% cow manure (CM100). Soil labile organic carbon components (particulate organic carbon, POC; microbial biomass carbon, MBC; readily oxidizing organic carbon, ROC) and carbon conversion related enzymes (α-D-glucosidase, AG; β-D-glucosidase, BG; Cellulase, CBH; Peroxidase, PER; Urease, UR) activity and other related indicators were measured.

(1) The content of SOC, POC and ROC in soil increased with the increase of the proportion of organic fertilizer applied. CM50CF50 had the highest MBC content, which was 139.7% higher than that under CK. In the non-fertilized area, compared with CK, the POC content under CF100, CM25CF75, CM50CF50 and CM75CF25 decreased by 32.8%, 28.4%, 21.6% and 14.7%, respectively. The ROC content under CM50CF50 and CM75CF25 treatments decreased by 31.5% and 17.4%, respectively. The content of labile organic carbon in fertilized area was significantly higher than that in non-fertilized area under the same treatment. (2) Compared with CK, the α-D-glucosidase activity under CM25CF75, CM50CF50, CM75CF25 and CM100 was increased by 87.7%, 68.4%, 278.1% and 331.6%, respectively. The β-D-glucosidase activity under CM25CF75 was the highest (39.00 µg·g^-1·h^-1). Urease activity first increased and then decreased with the increase of organic fertilizer application ratio. The α-D-glucosidase and urease activities of soil in the non-fertilized area were also significantly increased. (3) The combination of organic and inorganic application significantly increased the soil POC/SOC and carbon pool management index (CPMI) of the fertilization area, and the carbon pool management index under CM25CF75, CM50CF50, CM75CF25 and CM100 treatments increased by 19.7%, 38.3%, 56.2% and 103.5%, respectively. The carbon pool management index of organic and inorganic combined application in non-fertilized area decreased significantly. Soil carbon pool management index in fertilized area was significantly higher than that in non-fertilized area under the same treatment. (4) Correlation analysis and principal component analysis showed that there was a significant positive correlation between labile organic carbon components and α-D-glucosidase activity in the soil in the fertilization area, and the increase of organic fertilizer ratio contributed more to the increase of soil labile organic carbon. The effect of fertilization treatment on the fertilized area was greater than that on the non-fertilized area.

The effect of organic and inorganic combined application on soil improvement in fertilized area of apple orchard was greater than that in non-fertilized area. The combination of organic and inorganic application could increase the content of soil organic carbon and promote soil enzyme activity, which provided a theoretical basis for the sustainable management of soil ecological environment in apple orchard.