The invention discloses a multi-dimensional feature dynamic abnormal integral model based on a Markov-like model, which is applied to predictive analysis of illegal behaviors such as customs detour and card avoidance, and relates to the technical field of behavior analysis and probability theory. The method comprises the following steps: firstly, through a GPS, a camera, an electronic network police and an unmanned aerial vehicle, using an Internet of Things technology to obtain state information (including position, time, speed and man-vehicle state) of vehicles and personnel, starting from four directions of time characteristics, spatial path characteristics, man-vehicle state and specific behaviors, combining randomness of the man behaviors, and determining the state information of the vehicles and the personnel; and establishing an abnormal integral model of closing and card avoiding. On the basis of space and time characteristics, target state information is analyzed, meanwhile, two constraint conditions of'man-vehicle state 'and'specific behavior' are innovatively introduced, a transition probability matrix of the Markov model is dynamically updated according to abnormal characteristics, the gate-around and card-avoiding behaviors can be effectively predicted, and occurrence of the gate-around and card-avoiding behaviors can be stopped in time.

    本发明公开一种基于类马尔可夫模型的多维特征动态异常积分模型,应用于绕关避卡等违法行为预测分析,涉及行为分析和概率论技术领域。本发明包括以下步骤:首先通过GPS,摄像头,电子网警、无人机,运用物联网技术获取车辆及人员的状态信息(包括位置,时间,速度,人车状态),从时间特征、空间路径特征、人车状态和特定行为四个方向出发,结合人行为的随机性,建立绕关避卡的异常积分模型。本发明在空间和时间特征基础上,分析目标状态信息,同时创新性的引入“人车状态”及“特定行为”两项约束条件,根据异常特征动态更新类马尔可夫模型的转移概率矩阵,能够对绕关避卡行为进行有效预测,及时制止绕关避卡行为的发生。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Multidimensional feature dynamic abnormal integral model based on Markov-like model


    Weitere Titelangaben:

    一种基于类马尔可夫模型的多维特征动态异常积分模型


    Beteiligte:
    WANG QINGWANG (Autor:in) / HUANG JIANGBO (Autor:in) / SHEN TAO (Autor:in) / SONG JIAN (Autor:in) / TAO ZHIMIN (Autor:in) / LIU QUANJUN (Autor:in)

    Erscheinungsdatum :

    2023-03-07


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / 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



    A Mathematical Model of a Multidimensional Discrete Markov Random Process

    Trubin, I. S. | British Library Online Contents | 2005


    Multidimensional data based detecting method of traffic abnormal spots

    LI DAN / LI JIANYUAN / WANG HAO et al. | Europäisches Patentamt | 2015

    Freier Zugriff

    Multidimensional choice model transferability

    Koppelman, Frank S. / Pas, Eric I. | Elsevier | 1985


    Dynamic Hidden Markov Model for Metropolitan Traffic Flow Prediction

    Li, Zihan / Chen, Cailian / Min, Yang et al. | IEEE | 2020


    Path integral learning of multidimensional movement trajectories

    André, João / Santos, Cristina / Costa, Lino | BASE | 2013

    Freier Zugriff