轨道交通车站闸机智能识别研究

以南京地铁1号线为例,介绍其车站闸机传感器的分布位置。结合人体尺寸数据,采用事件及区域识别相结合的方法,分析了乘客的不同行为对传感器状态的影响。通过采用传感器设计、事件的处理及识别技术,以及与数据库的对比分析,使闸机系统能智能识别逃票尾随现象,减小了票务损失。通过在特定通道增加人脸识别功能,可实现有效分流,并提高闸机工作效率;通过ARIMA模型分析预测乘客高峰期,可合理配置资源。

Research on Intelligent Identification in Rail Transit Station Gate Machine

According to the distribution of ticket gate switch of Nanjing Metro Line 1,and combined with the data of human size,event and region recognition methods,the influence of different passenger behaviors on the sensor state is analyzed.On this basis,the sensor design,event recognition and handling techniques are adopted and compared with the data base,in order to reduce the fare evasion,trailing phenomenon and ticketing loss.Through adding the face recognition function in special lanes,gate machine working efficiency is increased and effective passenger division achieved.Also through passenger peak analysis by using the ARIMA model,reasonable allocation of resources is actualized.