Fully convolutional network is a powerful tool for per-pixel semantic segmentation/detection. However, it is problematic when coping with crack detection using partially accurate ground truths (GTs): the network may easily converge to the status that treats all the pixels as background (BG) and still achieves a very good loss, named “All Black” phenomenon, due to the unavailability of accurate GTs and the data imbalance. To tackle this problem, we propose crack-patch-only (CPO) supervised generative adversarial learning for end-to-end training, which forces the network to always produce crack-GT images while reserves both crack and BG-image translation abilities by feeding a larger-size crack image into an asymmetric U-shape generator to overcome the “All Black” issue. The proposed approach is validated using four crack datasets; and achieves state-of-the-art performance comparing with that of the recently published works in efficiency and accuracy.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    CrackGAN: Pavement Crack Detection Using Partially Accurate Ground Truths Based on Generative Adversarial Learning


    Beteiligte:
    Zhang, Kaige (Autor:in) / Zhang, Yingtao (Autor:in) / Cheng, Heng-Da (Autor:in)


    Erscheinungsdatum :

    01.02.2021


    Format / Umfang :

    3562482 byte




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Crack Detection Based on Generative Adversarial Networks and Deep Learning

    Chen, Gongfa / Teng, Shuai / Lin, Mansheng et al. | Springer Verlag | 2022



    Grid-based pavement crack analysis using deep learning

    Wang, Xianglong / Hu, Zhaozheng | IEEE | 2017


    Generative Adversarial Imitation Learning-Based Continuous Learning Computational Guidance

    Luo, Haowen / Lee, Chang-Hun / Li, Chaoyong et al. | IEEE | 2025


    Pavement crack detection using the Gabor filter

    Salman, M. / Mathavan, S. / Kamal, K. et al. | IEEE | 2013