摘要
本研究旨在设计和实现一套基于机器学习技术的列车运行安全预警系统,实时监控铁路调度集中系统(Centralized Traffic Control,CTC)和旅客服务系统之间的列车信息差异。该系统利用视频信号采集技术,结合深度学习模型,可实现对列车信息的智能识别与比对。通过物体检测模型,系统能够精准地定位到列车信息显示区域,而文字识别模型可以从这些区域中提取关键的列车运行信息,如车次和正晚点时间。系统的核心功能包括异常情况的实时报警、报警历史查询以及报警条件设置,以提高列车运行的安全性和旅客的出行体验。当系统检测到CTC和旅服系统之间的数据存在不一致时,会自动发出预警,提示综控人员及时进行信息纠正和信息同步,从而避免信息误差对旅客出行造成影响。
The aim of this study is to design and implement a train operation safety early warning system based on machine learning techniques to monitor the train information difference between the railway centralized traffic control(CTC)system and the passenger service system in real time.The system uses video signal acquisition technology combined with deep learning model to realize intelligent identification and comparison of train information.Through the object detection model,the system can accurately locate the train information display area,while the text recognition model is used to extract key train operation information from these areas,such as train number and positive delay time.The core functions of the system include real-time alarm of abnormal situations,alarm history query and alarm conditions setting to improve the safety of train operation and passenger travel experience.When the system detects data inconsistency between the CTC and the travel service system,it will automatically issue an early warning,prompting the comprehensive control personnel to correct and synchronize the information in time,so as to avoid the impact of information errors on passenger travel.
作者
张宇驰
王石宇
ZHANG Yuchi;WANG Shiyu(Shanghai Railsoft Information Technology Co.,Ltd.,Shanghai 200000,China)
出处
《信息与电脑》
2024年第13期126-129,共4页
Information & Computer
关键词
机器学习
物体检测
文字识别
machine learning
object detection
text recognition
