Unmanned aerial vehicle (UAV) patrol inspection has emerged as a predominant approach in transmission line monitoring owing to its cost-effectiveness. Detecting defects in transmission lines is a critical task during UAV patrol inspection. However, due to imaging distance and shooting angles, UAV patrol images often suffer from insufficient defect-related visual information, which has an adverse effect on detection accuracy. In this article, we propose a novel method for detecting defects in UAV patrol images, which is based on vision-language pretraining for transmission line (VLP-TL) and a progressive transfer strategy (PTS). Specifically, VLP-TL contains two novel pretraining tasks tailored for the transmission line scenario, aimimg at pretraining an image encoder with abundant knowledge acquired from both visual and linguistic information. Transferring the pretrained image encoder to the defect detector as its backbone can effectively alleviate the insufficient visual information problem. In addition, the PTS further improves transfer performance by progressively bridging the gap between pretraining and downstream defection detection. Experimental results demonstrate that the proposed method significantly improves defect detection accuracy by jointly utilizing multimodal information, overcoming the limitations of insufficient defect-related visual information provided by UAV patrol images.


    Access

    Download


    Export, share and cite



    Title :

    Transmission Line Defect Detection Based on UAV Patrol Images and Vision-language Pretraining


    Contributors:
    Zhang, Ke (author) / Zheng, Zhaoye (author) / Guo, Yurong (author) / Wang, Jiacun (author) / Yang, Jiyuan (author) / Xiao, Yangjie (author)

    Publication date :

    2024


    Type of media :

    Preprint


    Type of material :

    Electronic Resource


    Language :

    English




    Transmission line air patrol

    Carney, P.H. | Engineering Index Backfile | 1966


    Power transmission line patrol device

    LI RUIBO / WANG YI / LU XINGWANG et al. | European Patent Office | 2022

    Free access

    VILA: Learning Image Aesthetics from User Comments with Vision-Language Pretraining

    Ke, Junjie / Ye, Keren / Yu, Jiahui et al. | ArXiv | 2023

    Free access

    Power transmission line tower climbing patrol robot

    LIANG YING | European Patent Office | 2023

    Free access

    Power transmission line tower climbing patrol robot

    WANG LUPENG / ZHANG RENXIANG / SONG JIANFEI et al. | European Patent Office | 2021

    Free access