Railroad tracks need to be periodically inspected and monitored to ensure safe transportation. Automated track inspection using computer vision and pattern recognition methods has recently shown the potential to improve safety by allowing for more frequent inspections while reducing human errors. Achieving full automation is still very challenging due to the number of different possible failure modes, as well as the broad range of image variations that can potentially trigger false alarms. In addition, the number of defective components is very small, so not many training examples are available for the machine to learn a robust anomaly detector. In this paper, we show that detection performance can be improved by combining multiple detectors within a multitask learning framework. We show that this approach results in improved accuracy for detecting defects on railway ties and fasteners.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Deep Multitask Learning for Railway Track Inspection


    Beteiligte:
    Gibert, Xavier (Autor:in) / Patel, Vishal M. (Autor:in) / Chellappa, Rama (Autor:in)


    Erscheinungsdatum :

    2017-01-01


    Format / Umfang :

    2237755 byte




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Deep Multitask Learning for Railway Track Inspection

    Gibert, Xavier | Online Contents | 2017


    TrackNet - A Deep Learning Based Fault Detection for Railway Track Inspection

    James, Ashish / Jie, Wang / Xulei, Yang et al. | IEEE | 2018


    RAILWAY TRACK INSPECTION DEVICE

    HORIE MASAHIRO / KOBAYASHI HIROYUKI / TAKAHASHI YUSUKE et al. | Europäisches Patentamt | 2019

    Freier Zugriff

    Railway track inspection operation trolley

    LI PEIGANG / LAN CAIHAO / GUAN SHUGANG et al. | Europäisches Patentamt | 2020

    Freier Zugriff

    Railway track-based inspection device

    SONG MENG | Europäisches Patentamt | 2023

    Freier Zugriff