The invention relates to the technical field of traffic safety, and discloses a dangerous driving behavior assessment method and device. According to the method, firstly, traffic detection data are collected through a roadside detector, and driving behaviors are divided into normal driving, aggressive driving, drunk fatigue driving and distracted driving through hierarchical clustering. And then a deep learning multi-classification model is constructed according to a clustering result, the state of the driving behavior of the to-be-detected vehicle is monitored in real time, and whether a dangerous driving behavior exists or not is evaluated. The method can effectively cluster the distribution conditions and features of the four types of driving behaviors, and provides a reference basis for determining whether the driving behaviors belong to dangerous driving behaviors or not. Whether the driving behavior of the to-be-detected vehicle belongs to dangerous driving or not is evaluated by constructing the deep learning multi-classification model, the operation efficiency is high, and the accuracy and applicability are high. The driving behavior is judged by introducing the road traffic state data, so that the accuracy of the driving behavior judgment can be improved, and meanwhile, the road and road network levels are better planned and coordinated.

    本申请涉及交通安全技术领域,公开了一种危险驾驶行为的评估方法及装置。该方法首先通过路侧检测器采集交通检测数据,并通过层次聚类将驾驶行为分为正常驾驶、激进驾驶、酒后疲劳驾驶和分心驾驶。然后根据聚类结果构建深度学习多分类模型,实时监测待检测车辆的驾驶行为的状态,对是否存在危险驾驶行为进行评估。本申请能够有效聚类四类驾驶行为的分布情况和特征,为驾驶行为是否属于危险驾驶行为提供参考基准。通过构建深度学习多分类模型评估待检测车辆的驾驶行为是否属于危险驾驶,运算效率较高、准确度且适用性较强。通过引入道路通行状态数据,进行驾驶行为的判定,能够提高驾驶行为判定的准确度,同时,更好的从道路、路网级别统筹协调。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Dangerous driving behavior assessment method and device


    Weitere Titelangaben:

    一种危险驾驶行为的评估方法及装置


    Beteiligte:
    YANG YANG (Autor:in) / ZHANG ZHIXIANG (Autor:in) / LIU QIANG (Autor:in) / GUAN YONGSHENG (Autor:in)

    Erscheinungsdatum :

    2022-04-12


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06K Erkennen von Daten , RECOGNITION OF DATA / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    DANGEROUS DRIVING DETERMINATION DEVICE, DANGEROUS DRIVING DETERMINATION METHOD, AND DANGEROUS DRIVING DETERMINATION PROGRAM

    SATO TAKAYUKI / KASUYA JUNICHI / NAKAMURA MASAHIRO et al. | Europäisches Patentamt | 2021

    Freier Zugriff




    Dangerous driving behavior detection method and device

    ZHOU CHUNCAI / ZHOU YOU | Europäisches Patentamt | 2024

    Freier Zugriff