Traffic surveillance footage requires constant monitoring in order to detect deadly accidents and to react appropriately. However, it is tiresome and mistake prone to regularly supervise them with humans. A sophisticated and quick system for detecting road accidents is urgently needed due to the rise in fatalities from accidents on the roads. After accidents, a few crucial seconds frequently makes a lot of difference. There is a need for a system that can immediately recognize emergency situations and notify the police and hospital about them. This study proposes a real-time traffic incident detection and alarm system that is based on a computer vision technique, with the goal of achieving a penetrating impact in boosting the associated demand for road traffic security and safety. The vehicle detection and tracking model first assigned a distinct serial number (ID) to each vehicle in order to detect and track the movements of the vehicles. This was done using the YOLOv3 object detector and DeepSORT tracker with region of interest interception. The mean average precision (mAP) of this model was 98.2%.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    A Novel Approach for Accident Detection and Localization Using Deep Learning


    Contributors:


    Publication date :

    2023-11-23


    Size :

    1479218 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Accident Detection Using Deep Learning

    Yadav, Durgesh Kumar / Renu / Ankita et al. | IEEE | 2020


    Detection Of Railway Accident Risk Using Deep Learning Approach

    Arumugam, M. / Arun, G. / C, Dhanapal | IEEE | 2024


    Acoustic accident detection system through accident sound deep learning

    JOON HYUN KIM | European Patent Office | 2022

    Free access

    Acoustic accident detection system through accident sound deep learning

    KIM KYU HYUN | European Patent Office | 2022

    Free access

    Object Interaction-Based Localization and Description of Road Accident Events Using Deep Learning

    Thakare, Kamalakar Vijay / Dogra, Debi Prosad / Choi, Heeseung et al. | IEEE | 2022