The invention belongs to the field of intelligent traffic, and provides a malicious vehicle detection scheme based on traffic flow spatial-temporal characteristics. The scheme comprises five parts, wherein in the first part, the whole traffic network is divided into a plurality of sub-regions according to the intersection space correlation degree; in the second part, the roadside detection nodes in the subareas upload collected data information to corresponding fog servers, training is carried out through a graph attention network (GAT), and a model with spatial features is obtained; and the third part is that the cloud server collects all data processed in the fog server, and a global time feature is trained through a gating loop unit (A-GRU) with an attention mechanism to obtain a global space-time model. And the fourth part is construction of a reputation system, and reputation scores are updated according to the truth of the flow track. And the fifth part is malicious vehicle detection, wherein the system judges whether the vehicle is a malicious vehicle according to the reputation score of the vehicle.

    本发明属于智能交通领域,提供了一种基于交通流时空特征的恶意车辆检测方案。该方案包括五个部分,第一部分是根据交叉口空间关联度将整个交通路网划分为多个子区域。第二部分是子区域中路边检测节点将其收集到的数据信息上传到对应的雾服务器上,并且通过图注意力网络(GAT)进行训练,得到带有空间特征的模型。第三部分是云服务器将所有雾服务器中处理过的数据收集起来,通过带有注意力机制的门控循环单元(A‑GRU)对全局的时间特征进行训练,得到一个全局的时空模型。第四部分是信誉系统的构建,以流量轨迹真实度为依据对信誉分数进行更新。第五部分是恶意车辆检测,系统根据车辆的信誉分数来评判车辆是否为恶意车辆。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Malicious vehicle detection method based on traffic flow spatial-temporal characteristics


    Weitere Titelangaben:

    一种基于交通流时空特征的恶意车辆检测方法


    Beteiligte:
    OU YANGXIN (Autor:in) / GU KE (Autor:in)

    Erscheinungsdatum :

    14.07.2023


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES / 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




    MALICIOUS TRAFFIC DETECTION WITH ANOMALY DETECTION MODELING

    ACHLEITNER STEFAN / XU CHENGCHENG | Europäisches Patentamt | 2022

    Freier Zugriff

    MALICIOUS TRAFFIC DETECTION WITH ANOMALY DETECTION MODELING

    ACHLEITNER STEFAN / XU CHENGCHENG | Europäisches Patentamt | 2022

    Freier Zugriff

    Malicious traffic detection based on GWO-SVM model

    Liu, Lechao / Zhuang, Yan / Gao, Xufei | IEEE | 2022


    Short-Term Traffic Prediction considering Spatial-Temporal Characteristics of Freeway Flow

    Jiaqi Wang / Yingying Ma / Xianling Yang et al. | DOAJ | 2021

    Freier Zugriff