Unmanned Aerial Vehicles (UAVs) integrated with imaging technology offer aerial advantages and intelligent computer vision analysis capabilities, making significant contributions in various domains including geoinformatics, agriculture, environmental conservation, and infrastructure surveillance. This integration facilitates efficient and precise data assistance crucial for numerous sectors. However, challenges persist in balancing the scale and accuracy of target recognition, discrepancies in object recognition versus their actual sizes and issues like occlusions. To enhance the efficacy of UAV-based real-time vehicle monitoring, this study integrates an innovative Advanced Vehicle Detection Network (AVDN) into the fundamental network architecture of the YOLOv7-tiny model. This integration aims to enable swift, cost-effective, and accurate vehicle monitoring on roadways. Our findings indicate that the enhanced model achieves a mean Average Precision (mAP) of 56.14%, significantly surpassing the baseline YOLOv7-tiny model. The affordability and efficiency of this model underscore its practical application potential.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Employing YOLOv7-Tiny for Identifying Road Vehicles in Drone-Based Aerial Photography


    Contributors:
    Guo, Xia (author)


    Publication date :

    2024-06-28


    Size :

    6893707 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English





    Improved YOLOv7 Target Detection Algorithm Based on UAV Aerial Photography

    Zhen Bai / Xinbiao Pei / Zheng Qiao et al. | DOAJ | 2024

    Free access

    Efficient YOLOv7-Drone: An Enhanced Object Detection Approach for Drone Aerial Imagery

    Xiaofeng Fu / Guoting Wei / Xia Yuan et al. | DOAJ | 2023

    Free access

    YOLOv7-Tiny and YOLOv8n Evaluation for Face Detection

    Al Amoudi, Ibrahim / Ramli, Dzati Athiar | Springer Verlag | 2024