本发明公开了一种基于深度学习的斑马线礼让行人违法二次检测方法,属于图像识别领域。本发明的基于深度学习的斑马线礼让行人违法二次检测方法其基本步骤为,读入违法图片及其车道配置信息,对驾驶机动车的行人进行判断和剔除,识别人行道内行人和车辆,再对人行道中的行人以及车辆分区域计数,最终基于计数结果进行违法判定。本方法可有效提高违法数据筛选的效率,进而达到节省人力的效果。经过该方法对礼让行人违法数据二次识别,可为人工筛选出大量违法图片。

    The invention discloses a zebra crossing pedestrian violation secondary detection method based on deep learning, and belongs to the field of image recognition. In the zebra crossing pedestrian violation secondary detection method based on deep learning, illegal pictures and lane configuration information of the illegal pictures are read, pedestrians driving motor vehicles are judged and eliminated, pedestrians and vehicles in sidewalks are recognized, then the pedestrians and vehicles in the sidewalks are counted according to regions, and finally illegal judgment is conducted based on the counting result. According to the method, the illegal data screening efficiency can be effectively improved, so that the effect of saving manpower is achieved. Through the method, secondary identificationis carried out on the pedestrian violation data, and a large number of violation pictures can be manually screened out.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    基于深度学习的斑马线礼让行人违法二次检测方法


    Erscheinungsdatum :

    2021-03-02


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06K Erkennen von Daten , RECOGNITION OF DATA / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS