Utilizing radiative transfer models to simulate spectral data for improving the robustness and interpretability of leaf area index (LAI) inversion models is a major focus in current research. However, the application of point cloud data, which can directly reflect the physical structure of rice, remains underdeveloped. To bridge this gap, this study proposes a rice LAI inversion method that integrates a three-dimensional radiative transfer model with unmanned aerial vehicle (UAV) light detection and ranging (LiDAR) technology. Virtual rice canopy scenes were constructed using the LESS three-dimensional radiative transfer model to generate simulated point cloud data, while UAV LiDAR was employed to collect field-measured data. Pearson correlation analysis was used to select key features, leading to the construction of a feature subset comprising height features, vertical layer density features, and cumulative density features. Three machine learning models-ridge regression, support vector regression, and random forest-were trained and validated on simulated datasets, measured datasets, and a simulation-measurement hybrid dataset. The results show that the simulated data generated by the LESS model demonstrate strong consistency with measured point clouds in vertical distribution, confirming the validity of the simulation. Feature analysis indicates that the 90th percentile height (H90) and vegetation canopy volume (VCM) manifest the strongest correlations with LAI. Model comparison reveals that the random forest model performs best on the hybrid dataset, achieving a validation R² of 0.918 and RMSE of 0.787. The hybrid dataset strategy yields better fitting performance than using the measured dataset alone, effectively enhancing model generalization. In conclusion, the proposed method, which integrats three-dimensional radiative transfer modeling and UAV LiDAR technology, enables rapid and accurate estimation of rice LAI, providing a new technical pathway for precise monitoring of rice growth status and supporting high-throughput acquisition of crop phenotypic information within the framework of agricultural low-altitude economy.

Accurate quantitative prediction of crop yield is a crucial foundation for ensuring national food security and formulating smart agricultural decisions. During the rice growth cycle, the full heading stage is the most critical period for biomass accumulation, and its canopy spectral characteristics contain rich information on biochemical parameters and yield potential. This study explores the application advantages of combining multi-dimensional UAV remote sensing features with advanced machine learning algorithms for rice yield prediction at the full heading stage. Using a multispectral UAV to acquire images, this research systematically constructed six categories of feature sets, including red-green-blue (RGB), multispectral (MS), multispectral to hyperspectral (MSTPP), spatial texture (Texture), vegetation index (VI), and multi-dimensional fusion (Fusion). At the algorithmic level, eight machine learning algorithms were comprehensively evaluated, with a focus on comparing kernel methods represented by support vector regression (SVR) and various nonlinear ensemble models. The experimental results indicate that MS data exhibit outstanding prediction potential among single-dimensional features. The strategy of fusing spectral and texture information achieved excellent prediction stability, wherein the SVR model combined with multi-dimensional fusion features achieved an ideal fitting effect (coefficient of determination R²=0.828), calibrating the spectral signals impaired by canopy closure during the full heading stage, and effectively correcting the systematic estimation bias of single-dimensional models in extreme yield intervals. Furthermore, this experiment verified through the introduction of the Shapley additive explanations (SHAP) framework that the green normalized difference vegetation index (GNDVI) played a core driving role in yield prediction, objectively revealing the safeguard mechanism of population growth balance on the final yield. The approach of "multi-dimensional fusion combined with nonlinear algorithms and attribution analysis" explored in this study provides a feasible approach for constructing quantitative yield estimation models with both high robustness and agronomic interpretability.

As a core carrier of the low-altitude economy in the agricultural and rural sectors, agricultural unmanned aircraft have experienced explosive growth in recent years and have become a vital technological force driving agricultural modernization and rural revitalization. While the industrial scale has expanded dramatically and technological applications have undergone rapid iteration, the development of supporting regulations and standards has lagged behind. Based on a review of the development of domestic and international regulations and standards for agricultural unmanned aerial vehicles, this paper systematically analyzes the legal framework for airspace management, airworthiness certification, and personnel qualifications-as well as the current status of the four-tier standards framework including national standards, industry standards, local standards, and group standards. The findings indicate that China has initially established a regulatory framework with the Interim Regulation on the Flight Management of Unmanned Aircraft as the overarching legislation, and a standards architecture with the Guidelines for the Construction of the Low-Altitude Economy Standards System as the top-level design. However, shortcomings remain in areas such as application scenario coverage, component and subsystem requirements, and supporting methodological standards. Looking into the future, efforts should be directed toward advancing risk-based classification and hierarchical management, accelerating the expansion of application scenarios and the refinement of methods within the standards system, and building a digital intelligent supervision system, thereby providing institutional safeguards for the high-quality development of the agricultural low-altitude economy.

To improve the accuracy of identifying the maturity of tobacco leaves at different harvesting stages, this study systematically collected images of under-ripe, ripe, and over-ripe tobacco leaves from the upper, middle, and lower parts of the plant. We proposed an improved VGG-16-based deep learning method for tobacco leaf maturity recognition. This method leverages a pre-trained VGG-16 model as its foundation, incorporating transfer learning, fine-tuning, and convolutional layer feature fusion to enhance the model's performance in tobacco leaf maturity recognition tasks. Freezing the convolutional and subsequent layers, adding BN layers, and using the Adam optimizer further improved training efficiency, avoided over fitting, and enhanced the model's robustness and accuracy. Experimental results showed that the improved VGG-16 model had a high accuracy advantage in tobacco leaf maturity recognition tasks, with a test set accuracy of 99.7%, surpassing classical machine learning methods such as BP neural networks, support vector machines, original VGG-16 and VGG-19, AlexNet, and ResNet50. The model comprised 14 721 353 parameters, a model size of 58.9 M, and a single image recognition time of 0.024 9 seconds, demonstrating the advantages of low computational and storage resource requirements and rapid recognition speed. Further visual analysis of the recognition results of tobacco leaf images with different maturities using the Score-CAM algorithm revealed that the central region of the main vein of tobacco leaves served as a primary differentiating feature for tobacco leaves with different maturities across different stalk positions, providing key information for the recognition model, there by clarifying the chemical substance transformation patterns occurring during the maturity process of tobacco leaves from distinct plant sections. The improved VGG-16 deep learning model proposed in this study exhibits high accuracy and efficiency in identifying the maturity of tobacco leaves at different harvesting stages. It is poised to provide precise and effective decision support for tobacco harvesting and production. Future research will focus on exploring alternative feature fusion strategies and network structures to further improve the generalization ability and robustness of tobacco leaf maturity recognition across different production areas and years.

With the advancement of agriculture, the economic losses and food safety threats caused by major plant diseases and pests to agriculture and forestry are becoming increasingly severe. In 2024, China's main food crops face challenges from 22 types of diseases and pests, with an expected affected area of 155 411 khm² and a yield loss exceeding 1.5×10¹¹ t. This article provides an overview of the current status of crop disease and pest warning and emergency control technologies, where disease and pest warning technologies include the Internet of Things, remote sensing technology, and computer vision recognition technology; emergency control technologies cover the understanding of chemical communication mechanisms for the development of new pesticides and equipment, innovation in physical extermination and intelligent control products, utilization and product innovation of biological control resources, research on pest damage-type variation mechanisms and sustainable use of plant resistance varieties, research and development of pesticide resistance mechanisms and control technologies, as well as research and application technologies for the ecological regulation of diseases and pests. The aim is to achieve effective control of diseases and pests through interdisciplinary cooperation, while reducing environmental impact and promoting sustainable agricultural development. The article also discusses the application of disease and pest warning and emergency control technologies in the agricultural field, analyzes their important role in the development of agricultural science and technology, and summarizes the challenges faced in each field, such as the insufficient accuracy of unmanned aerial vehicle remote sensing monitoring technology, difficulties in sample collection for disease and pest image automatic recognition technology, the failure of chemical control to achieve precise application, the lagging production processes and market promotion of green formulations, and the incomplete construction of disease and pest warning systems. In the future, the application of Internet of Things technology to disease and pest warning should be accelerated, the construction of smart agriculture and disease and pest warning systems should be promoted, and research should be conducted on unmanned aerial vehicle remote sensing monitoring technology, disease and pest image automatic recognition technology, chemical control technology, and biological control technology.

In China, walnut kernel grading predominantly relies on mechanical and manual methods, which suffer from high economic costs, low efficiency, and poor accuracy, severely limiting the standardization and economic value of walnut kernel products. To achieve automated and intelligent walnut kernel grading, this study proposes a deep learning-based intelligent grading method. An image acquisition device was utilized to capture walnut kernel images across six grades under three distinct background colors: pure white, light green, and conveyor belt. The Python OpenCV library was employed for data augmentation to expand the dataset, resulting in a VOC format dataset of 12 246 images, with training, testing, and validation sets allocated in an 8：1：1 ratio. To select the most suitable model for walnut kernel grading, two models were constructed based on the YOLOv5 and YOLOv7 object detection networks, respectively. Both models were trained using pre-trained weights, achieving average grading accuracies of 87.83% and 91.16% for YOLOv5 and YOLOv7, respectively. During validation, YOLOv7 exhibited mispredictions, prompting the integration of two attention mechanisms, ECANet and CBAM, to refine the model. The improved models demonstrated enhanced performance, with the YOLOv7+CBAM model achieving the highest average accuracy (94.5%) and F1-score (90.2%). Compared to the original YOLOv7, the upgraded model increased average accuracy and F1-score by 3.34% and 5.9%, respectively, while adding only 2ms to inference time. To validate practical feasibility, a recognition system platform was developed for walnut kernel classification testing. Four experimental groups were designed, each containing 120 walnut kernels (20 per grade). The YOLOv7+CBAM model achieved an average recognition accuracy of 91.63%, demonstrating robust grading capabilities for walnut kernel appearance quality. This study provides a reference for intelligent walnut kernel grading systems.

To address the degradation of power generation efficiency caused by soiling accumulation in photovoltaic (PV) power plants, as well as the localization challenges faced by cleaning unmanned aerial vehicles (UAVs) due to limitations in edge computing power and complex background interference, this paper proposes an edge-computed cleaning UAV solution based on the Jetson Orin Nano platform. Within the YOLO11n framework, a lightweight photovoltaic panel soiling detection algorithm designated as GOP-YOLO11n is developed. By introducing GroupDC structure, the C3k2 modules in the backbone network undergo lightweight reconstruction, minimizing computational overhead while enhancing the feature representation of small targets against complex backgrounds. Furthermore, an OptimizedDC multi-scale feature extraction module is embedded into the feature fusion network to improve the perception of multi-scale targets, such as tiny bird droppings and elongated water stains. In the detection head, an upsampling and feature aggregation block (UP-Block) is implemented to strengthen multi-pixel pattern aggregation. For the experiments, a visible-light image dataset of PV panels captured by UAVs, yielding 4 190 valid imagespartitioned into training, validation, and test sets at an 8：1：1 ratio, with input resolution standardized to 960 pixels × 960 pixels. Under identical training and testing configurations, GOP-YOLO11n achieved a Precision of 91.1%, a Recall of 85.8%, a mAP@50 of 92.5%, and a mAP@50：0.95 of 76.8%, outperforming the baseline YOLO11n model by 2.9, 3.4, 1.8, and 4.4 percentage points, respectively. The model parameters were reduced from 2.582 M to 2.068 M, with a floating-point operation count of 6.7×10⁹. Moreover, hardware deployment tests using TensorRT FP16 on the Jetson Orin Nano Super 8 G platform showed an inference speed of 114.7 f/s, a single-frame latency of 8.7 ms, and a GPU memory footprint of 1 685 MB. The results indicate that the proposed method effectively reconciles the need for high-accuracy multi-scale soiling detection in complex scenes with the constraints of lightweight edge-side real-time inference, thereby providing robust visual perception support for soiling recognition, localized cleaning strategies, and the intelligent operation and maintenance (O&M) of PV cleaning UAVs.

As the low-altitude economy joins the ranks of national strategic emerging industries, the integration of low-altitude agricultural technology and equipment into agricultural production continues to deepen, serving as a cornerstone for promoting agricultural modernization and intelligent agriculture. This paper systematically evaluates the development status and future trends of low-altitude agricultural technology and equipment in China, focusing on policy frameworks, key technologies, and typical application scenarios. First, from the perspective of aerial platforms and intelligent control, the paper summarizes the current development of multi-rotor, single-rotor, and hybrid unmanned aerial vehicles (UAVs), and traces the evolution of key technologies in power systems, flight control algorithms, and navigation positioning. Concurrently, the progress of intelligent control technologies is outlined with an emphasis on low-altitude communication, cloud-based management, and multi-UAV coordination. Second, regarding low-altitude sensing and precision operations, the paper highlights the applications of UAV-based remote sensing in crop monitoring, alongside the technological pathways of precision spraying and broadcast sowing, identifying variable-rate operations and intelligent control as pivotal research frontiers. Finally, by examinging the inspection and evaluation systems and standardization frameworks of low-altitude agricultural equipment, the paper identifies existing challenges in operational specifications, quality evaluation, and systematic standards. The analysis reveals that although low-altitude agricultural technology offers significant advantages in operational efficiency and resource utilization, it remains constrained by insufficient equipment adaptability, limited operational stability in complex environments, inadequate multi-source information fusion and weak intelligent decision-making capabilities, and incomplete standard systems. Future research should prioritize multi-source sensing fusion, intelligent control, and data-driven decision-making technologies, while improving standard and evaluation systems, to promote the development of low-altitude agricultural technology toward precision, intelligence, and large-scale applications. This study provides a reference for research, equipment development, and policymaking in the field of low-altitude agriculture.

The development of the low-altitude economy is reshaping the technological functions, organizational forms, and regulatory demands of agricultural aircraft. Represented by agricultural drones, such aircraft have evolved from seasonal plant protection equipment into low-altitude production units integrating field operations, crop and farmland sensing, disaster response, and socialized service provision. As application boundaries continue to expand, the risks associated with agricultural low-altitude operations have also shifted from conventional flight safety concerns to a more comprehensive governance problem in which flight safety, operational quality, environmental externalities, and liability traceability are mutually intertwined. Against this background, a comparative analytical framework was developed in this study centered on airspace accessibility, operational authorization, regulatory visibility, operational compliance, and enforcement traceability. And the governance regimes for agricultural low-altitude operations across major jurisdictions, including China, the United States, the European Union, Australia and New Zealand, were systematically reviewed. In addition, a regulatory release density indicator was introduced, together with publicly available data on registered operators, registered unmanned aircraft, and remote pilots, to enable comparison of the relationships between diverse regulatory supply forms and large-scale operational capacity. The findings showed that major jurisdictions have developed three broad governance pathways: a strong-foundation model, a high-efficiency authorization model, and a risk-assessment-and-service model. In agricultural scenarios, institutional effectiveness depends not on the sheer number of individual rules, but on whether airspace access, digital identification, operational oversight, and evidentiary record retention can be integrated into an executable and verifiable engineering closed loop. Moreover, the concurrent expansion of common application scenarios, including production operations, remote sensing monitoring, emergency support, and cross regional socialized services, has further intensified the demand for differentiated authorization, tiered regulation of payload release, and cross regional collaborative oversight. On this basis, the study proposed an engineering-oriented compliance pathway centered on mission-based classification, the coupling of authorization and traceable records, tiered control of payload release, and agriculture-oriented continuing airworthiness management. The results may provide a useful reference for improving Chinese governance system of the agricultural low-altitude economy and for promoting the standardized application of agricultural drones.

Real-time and accurate monitoring of the winter wheat nitrogen nutrition index (NNI) is of great significance for optimizing field water and fertilizer management and reducing agricultural non-point source pollution. To address the issues of modal confusion in existing deep networks when processing multi-source UAV remote sensing data, as well as the loss of two-dimensional spatial topological information during temporal modeling, this study proposes a dual-branch multi-source feature fusion memory network (DMFM-Net). Multispectral images and canopy structural data of winter wheat at the tillering, jointing, and booting stages were collected by UAV. An enhanced CBAM-ResNet50 dual-branch backbone was employed to separately extract spectral features and canopy structural features. The spatially preserved feature tensors were then directly input into a convolutional long short-term memory (ConvLSTM) network for temporal modeling. The results show that: (1) Compared with the traditional serial network (CBAM-ResNet-LSTM) and the single-branch network (CBAM-ResNet-ConvLSTM), DMFM-Net effectively overcomes the noise interference of multimodal heterogeneous data and achieves the best inversion performance at all key growth stages. The coefficients of determination (R²) of the model at the tillering, jointing, and booting stages reach 0.76, 0.86, and 0.82, respectively, with root mean square errors (RMSE) as low as 0.11, 0.10, and 0.14. (2) Ablation experiments further confirm the necessity and scientific validity of the proposed architectural design. The ConvLSTM model lacking a deep spatial encoder shows a significant decrease in R² of 0.21-0.24. Moreover, compared with the static spatiotemporal feature extraction of 3D-ResNet, DMFM-Net, by virtue of its ConvLSTM module covering a long-range dynamic memory mechanism across the whole growth period, improves the R² of the inversion models at each stage by 0.07-0.16. (3) The field-scale spatial distribution maps generated by the optimal model accurately reconstruct the local spatial heterogeneity and temporal evolution patterns of winter wheat nitrogen nutritional status under different nitrogen application gradients. In summary, this study provides a new theoretical paradigm and technical support for multimodal time-series dynamic monitoring of crop nutritional status in complex field environments.