The drastic increase in vehicular traffic has led to an upsurge in road accidents. A crucial factor that significantly influences the outcome of an accident is the response time. Delayed response times can lead to fatalities, the main causes being inaccessible medical attention and traffic congestion. In this paper, we propose an effective system to minimize the response time by integrating Accident Detection and Smart Traffic System. Our Accident Detection module uses a Convolutional Neural Network (CNN) for real-time analysis of camera feeds in pre-identified accident-prone areas. Upon detecting an accident, the system promptly sends a notification to the closest hospital via the Telegram app. The notification includes a photo of the accident, date, time and precise location. The Smart Traffic system employs Yolov3 to detect and count vehicles in each lane at a junction. The lane-level microcomputers continuously communicate the vehicle counts to a central unit, which dynamically controls the traffic signals based on the traffic density. Our adaptive approach minimizes the time taken to notify emergency services as well as synchronizes traffic, thereby ensuring faster respite to accident victims.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Accident Detection & Smart Traffic System using CNN


    Beteiligte:


    Erscheinungsdatum :

    19.10.2024


    Format / Umfang :

    591958 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    DETECTING AND PROCESSING SYSTEM OF TRAFFIC ACCIDENT USING SMART STAND

    KIM YOUNG SOO | Europäisches Patentamt | 2016

    Freier Zugriff

    Smart Vehicle Accident Detection System

    Kumar, A Sunil / N, Amith / A, Jagadeesan et al. | IEEE | 2021


    Intelligent Traffic Accident Detection System Using Surveillance Video

    Sun, Pengfei / Liu, Qinghe | Springer Verlag | 2022