An improved YOLOv5 network with SPP and CBAM added to the YOLOv5 network is proposed to identify airport service lane damages accurately, quickly, and completely on the basis of ground-penetrating radar technology. First, the structure and performance of YOLOv5 are described in detail, while the common hidden defects are marked in combination with the actual GPR image data acquired at the airport. The improved network achieved a 6.2% improvement in network detection accuracy, a 5.7% improvement in recall rate, and a 5.1% improvement in mAP compared to the original one. These findings demonstrated the high accuracy and feasibility of the proposed optimized network for the detection of hidden defects in airport service lanes.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    An improved Yolov5-based algorithm for pavement disease detection in airport flight areas


    Contributors:
    Zhang, Kun (editor) / Lorenz, Pascal (editor) / Wang, Sibo (author) / Hao, Shougang (author) / Zhang, Ping (author) / Cao, Tie (author) / Shao, Liming (author) / Li, Youyang (author)

    Conference:

    International Conference on Mechatronics and Intelligent Control (ICMIC 2024) ; 2024 ; Wuhan, China


    Published in:

    Proc. SPIE ; 13447 ; 134473D


    Publication date :

    2025-01-16





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Lightweight pear detection algorithm based on improved YOLOv5

    Hu, Xiaomei / Zhang, Yunyou / Chen, Yi et al. | SPIE | 2023


    Multi-target Detection in Airport Scene Based on Yolov5

    Kun, Yan / Man, Hua / Yanling, Li | IEEE | 2021



    Mixed Traffic Flow Detection Algorithm Based on Improved YOLOv5

    Wang, Zhiyong / Lv, Bin / Chen, Qixiang | Springer Verlag | 2025


    Real-Time Detection of Voids in Asphalt Pavement Based on Swin-Transformer-Improved YOLOv5

    Zhang, Bei / Cheng, Haoyuan / Zhong, Yanhui et al. | IEEE | 2024