With the rapid growth of motor vehicles and highway construction, car accident becomes a major problem. Fast detection of an accident can save human life, reduce rescue time and public resources. In this paper, we propose a fast and implementable model for highway accident detection based on big data method and mobile operators' call data. To avoid the cost of infrastructure construction, sensor installation and data collection barriers between different departments, this model only uses the mobile operators' measurement report (MR) data to build the model, without any additional installation of traffic sensors, cameras, or any other detection equipment. This model creates a fingerprint map for a road with direction and location labels. Then this model uses the logistic regression (LR) algorithm to determine a passing mobile vehicle's location based on the map and MR data, alarm with different degree raised after a decision rule include the vehicle's location, velocity and velocity reduction, as well as emergency alarm suggestion for the following vehicles.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    A Low Cost and Easy Implement Highway Accident Detection Model Based on Big Data


    Beteiligte:
    Song, Chuntao (Autor:in) / Wang, Yong (Autor:in) / Zhang, Shiyan (Autor:in) / Xu, Lexi (Autor:in)


    Erscheinungsdatum :

    2019-10-01


    Format / Umfang :

    509973 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Accident cost saving and highway attributes

    Forkenbrock, David J. / Foster, Norman S. J. | Online Contents | 1997


    Prioritizing Highway Accident Sites: A Data Envelopment Model

    Cook, W. / Kazakov, A. / Persaud, B. | British Library Conference Proceedings | 1997