Search

Object detection in unmanned aerial vehicle (UAV) images presents challenges such as high altitudes, small object sizes, and complex backgrounds. Additionally, many deep learning object detection algorithms require substantial computational resources, making them difficult to deploy on embedded devices with limited memory and processing power, which affects the effectiveness of drones in task execution. To tackle these issues, we propose the LightUAV-YOLO algorithm which is a lightweight object detection algorithm for UAVs based on YOLOv8n. We modified the neck structure of YOLOv8, enhancing the network’s capability to detect small objects. To further optimize features fusion at different scales, we designed the orthogonal feature enhancement module (OFEM) which replaces simple concatenation for better feature representation. We also designed the local attention module (LAM) to effectively filter out irrelevant interference. The module enables our model better focus on important areas and further enhancing the model’s robustness. Results demonstrate that our proposed LightUAV-YOLO algorithm achieves a 6.4 and 3.9% improvement in mAP50 and mAP50:95, respectively, on the VisDrone test dataset compared to the YOLOv8-nano. Meanwhile, the model maintains a low parameter count and computational complexity. Furthermore, we conducted extensive experiments on the UAVDT dataset, and our method consistently exhibited favorable results. This model not only meets accuracy requirements but also considers the lightweight requirements. [ABSTRACT FROM AUTHOR]

The incorporation of difluoromethylene (CF2) group into chemical molecules often imparts desirable properties such as lipophilicity, binding affinity, and thermal stability. Consequently, the increasing demand for gem-difluoroalkylated compounds in drug discovery and materials science has continued to drive the development of practical methods for their synthesis. However, traditional synthetic methods such as deoxofluorination often confront challenges including complicated substrate synthesis sequences and poor functional group compatibility. In this context, we herein report a metal electron-shuttle catalyzed, modular synthetic methodology for difluoroalkylated compounds by assembling two C(sp3) fragments across CF2 unit in a single step. The approach harnesses a difluoromethylene synthon as a biradical linchpin, achieving the construction of two C(sp3)-CF2 bonds through radical addition to two different π-unsaturated molecules. This catalytic protocol is compatible with broad range of coupling partners including diverse olefins, iminiums, and hydrazones, supporting endeavors in the efficient construction of C(sp3)-rich difluoroalkylated molecules. The difluoromethylene subunit can confer advantageous biological properties to compounds of interest, and the incorporation of this moiety in mild and selective conditions is an active pursuit in synthetic organic chemistry. Here, the authors append a CF2 biradical synthon to two carbon units via cobalt catalysis. [ABSTRACT FROM AUTHOR]

Autonomous Underwater Vehicle (AUV) have become an important tool for humans to explore the ocean. The evolution of AUV is trending towards clustering. At present, research on the intelligent control strategy of AUV mostly focuses on the path following and trajectory tracking. Meanwhile, few studies concentrate on the control of an AUV in emergency situations. In the scenario of network attack, an emergency surfacing is necessary. In case of emergency, the loss of AUV can be reduced by timely and stable surfacing. The superiority of reinforcement learning (RL) in AUV control has been proven by many studies. In this work, we used an improved deep reinforcement learning (DRL) method based on deep deterministic policy gradient (DDPG) to solve the problem of AUV surfacing control in emergency situations. The method introduces expert experience data for pre-training and changes the update mechanism of the actor network, thereby improving the convergence rate and stability of the algorithm. Furthermore, we simulate the AUV surfacing control in five typical emergency situations, including partial thruster damage and rudder-jamming. Experimental results indicate that our method enables AUV to choose shorter paths in emergency situations, surfacing to the target area. Moreover, the method converges and maintains a stable reward function after training up to 900 episodes, showing faster convergence and stable performance. [ABSTRACT FROM AUTHOR]

The addition of ectomycorrhizal fungi (ECMF), beneficial rhizosphere microorganisms, to the soil can promote plant growth and resistance. Here, Populus davidiana × Populus bolleana tissue culture seedlings were grown for 3 months in soils inoculated with one of the species, then seedlings were assessed for mycorrhizal colonization rate and growth, physiological and root traits. Suillus luteus and Populus involutus each formed ectomycorrhizal associations with the seedlings. Seedling height, ground diameter, biomass, and leaf area were significantly greater after treatment with ECMF than in the non-inoculated controls. Treatment improved all physiological and root variables assessed (chlorophylls and carotenoids, cellulose, and soluble sugars and proteins; root length, surface area, projected area, mean diameter, volume, number of root tips). Seedlings inoculated with S. luteus outperformed those inoculated with P. involutus. [ABSTRACT FROM AUTHOR]

Wheat plants showing yellowing and mosaic in leaves and stunting were collected from wheat fields in Henan Province, China. Analysis of these plants by transmission electron microscopy showed that they contained two types of filamentous virus-like particles with a length of 200-500 nm and 1000-1300 nm, respectively. RNA-seq revealed a coinfection with wheat yellow mosaic virus (WYMV) and an unknown wheat-infecting virus. The genome of the unknown virus is 8,410 nucleotides long, excluding its 3' poly(A) tail. It has six open reading frames (ORFs). ORF1 encodes a putative viral replication-associated protein (Rep), and ORFs 2, 3, and 4 encode the triple gene block (TGB) proteins. ORFs 5 and 6 encode the capsid protein (CP) and a protein with unknown function, respectively. Phylogenetic analysis showed that this novel virus is evolutionarily related to members of the subfamily Quinvirinae, family Betaflexiviridae. It is, however, distinct from the viruses in the currently established genera. Based on the species and genus demarcation criteria set by the International Committee on Taxonomy of Viruses (ICTV), we tentatively name this novel virus "wheat yellow stunt-associated betaflexivirus" (WYSaBV), and we propose it to be a member of a new genus in the family Betaflexiviridae. [ABSTRACT FROM AUTHOR]