In vision-based environment perceptive system of urban rail transit, cameras installed in the front of the train can assist people to identify obstacles on rails. Varying environment makes small targets like pedestrians and bags hard to identify, thus the real-time and accuracy performance of the detection need to be improved. Inspired by the achievements of Single Shot Multibox Detection (SSD) successfully applied on images recognition in recent years, we presented an obstacle detection algorithm which consists of two steps: main network and feature fusion. In the first part, the input image is converted into multi-scale feature maps based on the Residual Neural Network. Next, a series of convolution layers are added to extract features, and the network outputs a confidence score and bounding boxes for possible obstacles. Experiments showed that the proposed method can detect obstacles in various environment. Compared with traditional object detection algorithms and other deep learning algorithms, it runs at a faster detection speed (26 frames per second, FPS) and higher detection accuracy (91.61% mean average precision, mAP) on our self-made dataset.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Real-time Obstacle Detection Over Rails Using Deep Convolutional Neural Network


    Beteiligte:
    Xu, Yuchuan (Autor:in) / Gao, Chunhai (Autor:in) / Yuan, Lei (Autor:in) / Tang, Simon (Autor:in) / Wei, Guodong (Autor:in)


    Erscheinungsdatum :

    2019-10-01


    Format / Umfang :

    2628527 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Real‐time CVSA decals recognition system using deep convolutional neural network architectures

    Juan Yépez / Riel Castro‐Zunti / Younhee Choi et al. | DOAJ | 2021

    Freier Zugriff

    Real‐time CVSA decals recognition system using deep convolutional neural network architectures

    Yépez, Juan / Castro‐Zunti, Riel / Choi, Younhee et al. | Wiley | 2021

    Freier Zugriff

    Emergency Vehicle Detection Using Deep Convolutional Neural Network

    Haque, Samiul / Sharmin, Shayla / Deb, Kaushik | Springer Verlag | 2022


    Real-time Pedestrian Detection using Convolutional Neural Network on Embedded Platform

    Hu, Jun / Liu, Wei / Cheng, Shuai et al. | British Library Conference Proceedings | 2016


    HDRT: Helmet Detection System using Convolutional Neural Network in Real Time

    U, Vasanthakumar G / Kumar K, Dheeraj / K, Pavan Reddy et al. | IEEE | 2022