This paper demonstrates a solution to detecting defects of hook-shaped fasteners by using a 3D system. Railway inspection including detection of defective fasteners is an important task in railway maintenance, and frequently conducted to prevent safety and operational problems. A new fastener detection method based on the 3D laser imaging system is proposed to overcome the shortcomings of traditional detection methods such as illumination inequality and stability issues in detection results. The line laser used in the 3D sensors has the excellent quality of good brightness, monochromaticity, and coherence. When the line laser irradiates vertically on the fasteners, the camera captures the beam which is formed by the fasteners diffuse reflection, so as to acquire the original 3D data of the fasteners. Several computer vision methods have been reviewed for fastener inspection applications. However, the methods are based on 2D images with unstable detection rate affected by image quality. In this paper, based on the collected 3D information of fasteners, characteristics of broken or missing fasteners are modeled and an evaluation method is proposed to detect partially broken or missing fasteners. In order to locate fasteners correctly, the prior-knowledge is used to verify the location of fasteners. Moreover, the clip sub-images are picked up to extract features for analysis. Then virtual broken fasteners are created to balance the number of intact and broken fastener images. Using the height gradients oriented histogram method and a combination of two classifiers, the system described in this paper can accurately inspect the defective hook-shaped fasteners. Preliminary results show that the proposed method can produce highly accurate evaluation data in a computationally efficient manner.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Railway Fastener Detection Method Based on 3D Images


    Beteiligte:
    Dai, Xianxing (Autor:in) / Peng, Yi (Autor:in) / Wang, Kelvin C. P. (Autor:in) / Yang, Enhui (Autor:in) / Li, Josh Q. (Autor:in) / Ding, Shihai (Autor:in)

    Kongress:

    First International Conference on Rail Transportation 2017 ; 2017 ; Chengdu, Sichuan Province, China


    Erschienen in:

    ICRT 2017 ; 938-946


    Erscheinungsdatum :

    13.03.2018




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Railway fastener defect form rapid detection vehicle

    LIU LINYA / WU SONGYING / CUI WEITAO | Europäisches Patentamt | 2021

    Freier Zugriff

    Railway fastener operation vehicle

    LEI ZHANGWEN / CHEN RONG / LIU YONG et al. | Europäisches Patentamt | 2020

    Freier Zugriff

    Rail fastener detection of heavy railway based on deep learning

    Yuan Cao / Zihao Chen / Tao Wen et al. | DOAJ | 2023

    Freier Zugriff

    Mask R-CNN Architecture Based Railway Fastener Fault Detection Approach

    Yilmazer, Merve / Karakose, Mehmet | IEEE | 2022


    Railway fastener base plate settlement volume rapid detection device

    ZENG ZHIPING / GUO WUJI / YE MENGXUAN et al. | Europäisches Patentamt | 2021

    Freier Zugriff