A rail-vehicle mounted system (10, Fig.1) comprises a sensor unit 20 directed at a track 32 ahead of the vehicle 30 in the direction of motion. The sensor data may be LIDAR, SONAR or video data, and is analysed in real time to indicate track conditions. E.g. image analysis may detect objects on the track, a direction in which the track is proceeding, a status of railway points 34 or a track defect. If an object is detected, an emergency system may sound an alarm or apply brakes. The detected track direction may be used to anticipate track layout, e.g. speed modulation. Once the train’s future path is determined (e.g. using the points setting indicated by point tips (switch plates) 38, Fig.2C) a directional alarm can protect a region in the direction it will travel while ignoring other directions. The information may be communicated to other nearby systems or a central information store. Information received from them may be confirmed or updated.


    Access

    Download


    Export, share and cite



    Title :

    Real-time computer vision-based track monitoring


    Contributors:

    Publication date :

    2022-09-21


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    English


    Classification :

    IPC:    B61L Leiten des Eisenbahnverkehrs , GUIDING RAILWAY TRAFFIC



    Real-time computer vision-based track monitoring

    European Patent Office | 2021

    Free access

    REAL-TIME COMPUTER VISION-BASED TRACK MONITORING

    MARSHALL ARCHIE | European Patent Office | 2023

    Free access

    Computer vision based real-time pixel-level railroad track components detection system

    QIAN YU / GUO FENG | European Patent Office | 2025

    Free access

    COMPUTER VISION BASED REAL-TIME PIXEL-LEVEL RAILROAD TRACK COMPONENTS DETECTION SYSTEM

    QIAN YU / GUO FENG | European Patent Office | 2021

    Free access

    Real Time Computer Vision

    B. Kallemov | NTIS | 2021