During the flowering period of mango trees, pests often hide in the inflorescences to suck sap, affecting fruit formation. By accurately detecting the number and location of mango inflorescences in the early stages, it can help target-specific spraying equipment to perform precise pesticide application. This study focuses on mango panicles and addresses challenges such as high crop planting density, poor image quality, and complex backgrounds. A series of improvements were made to the YOLOv8 model to enhance performance for this type of detection task. Firstly, a mango panicle dataset was constructed by selecting, augmenting, and correcting samples based on actual agricultural conditions. Second, the backbone network of YOLOv8 was replaced with FasterNet. Although this led to a slight decrease in accuracy, it significantly improved inference speed and reduced model parameters, demonstrating that FasterNet effectively reduced computational complexity while optimizing accuracy. Further, the GAM (Global Attention Module) attention mechanism was introduced as an attention module in the backbone network to enhance feature extraction capabilities. Experimental results indicated that the addition of GAM improved the average precision by 2.2 percentage points, outperforming other attention mechanisms such as SE, CA, and CBAM. Finally, the model’s bounding box localization ability was enhanced by replacing the loss function with WIoU, which also accelerated model convergence and improved the mAP@.5 metric by 1.1 percentage points. Our approach demonstrates a discrepancy of less than 10% compared to manual counted results.

Picking fruits from tall fruit trees manually is laborious and inefficient. Rotary-wing drones, a low-altitude carrier platform, can enhance the picking efficiency for tall fruit trees when combined with picking robotic arms. However, during the operation of rotary-wing drones, the wind field changes dramatically, and the center of gravity of the drone shifts at the moment of picking, leading to poor aerodynamic stability and making it difficult to achieve optimized attitude control. To address the aforementioned issues, this paper constructs a drone and wind field testing platform and employs the Lattice Boltzmann Method and Large Eddy Simulation (LBM-LES) algorithm to solve the high-dynamic, rapidly changing airflow field during the transient picking process of the drone. The aerodynamic structure of the drone is optimized by altering the rotor spacing and duct intake ratio of the harvesting drone. The simulation results indicate that the interaction of airflow between the drone’s rotors significantly affects the stability of the aerodynamic structure. When the rotor spacing is 2.8R and the duct ratio is 1.20, the lift coefficient is increased by 11% compared to the original structure. The test results from the drone and wind field experimental platform show that the rise time ($t_r$) of the drone is shortened by 0.3 s, the maximum peak time ($t_p$) is reduced by 0.35 s, and the adjustment time ($t_s$) is accelerated by 0.4 s. This paper, by studying the transient wind field of the harvesting drone, clarifies the randomness of the transient wind field and its complex vortex structures, optimizes the aerodynamic structure of the harvesting drone, and enhances its aerodynamic stability. The research findings can provide a reference for the aerodynamic optimization of other types of drones.

In UAV-based remote sensing, accurate and efficient image mosaicing is crucial for achieving real-time monitoring. Traditional cloud-centric processing paradigms, however, face core scientific challenges such as high latency, bandwidth bottlenecks, and limited autonomy, making them inadequate for dynamic, real-time scenarios. To address these issues, this paper proposes an edge-computing-enabled UAV image mosaicing system. The system consists of a UAV remote sensing platform and an edge computing terminal, with the core being our novel B-SIFT-ILS algorithm. The algorithm first uses geographic coordinates for unified registration, constructs a Gaussian scale space for multi-resolution representation, and then precisely locates extrema in the Difference of Gaussian (DoG) space using a 3D quadratic function. A BANSAC algorithm is subsequently employed to refine feature points and extract stable SIFT features, and finally, Iterative Least Squares (ILS) are used to achieve seamless mosaicing. Experimental results demonstrate that, compared with classical RANSAC, the proposed method achieves superior feature sampling accuracy (rotation: 0.879, translation: 0.877) and lower latency. The ILS-based smoothing stage effectively eliminates noise and ghosting without introducing gradient reversal, performing comparably to deep learning methods while significantly outperforming direct averaging and Gaussian approaches. On the NVIDIA Jetson Orin NX edge terminal, a single processing instance requires only 1124 ms, highlighting its strong potential for real-time, low-latency, and autonomous mosaicing tasks. Future research will focus on extending the approach to non-planar terrains and implementing adaptive parameter tuning for the BANSAC algorithm.