摘要
对重点区域周界入侵实现自动、准确、实时的报警日益得到重视。笔者介绍了一种利用支持向量机进行模式分类的方法,该方法对周界现场振动信号进行处理,从而实现周界入侵事件的智能识别,提高了系统的准确性和稳定性。高铁光纤周界入侵报警系统应用在国内高铁线路的周界安全监测中,现场实验测试和使用表明:对于高铁非法入侵行为,系统能准确判断并及时报警,系统还能够屏蔽风雨雪等非入侵事件。系统的推广应用将具有显著的经济效益和重大的社会意义。
It attracts more and more attentions to automatically give a prompt alarm for perimeter intrusion in important regions.The author introduces a method of pattern recognition which can be used for signal processing of the vibration signals of perimeter.By means of this artificial intelligence method,the system accuracy and stability can be improved.The intrusion alarm system of highspeed railway optical fiber is applied to the perimeter safety monitoring of high-speed rail line in China.The application shows that for high-speed rail illegal intrusion,the system can judge an intrusion and give an alarm.The system can also shield wind and snow and other non-invasive events.The application of the system will have significant economic benefits and important social significance.
作者
张国全
Zhang Guoquan(Harbin-Dalian Railway Passenger Dedicated Line Co.,Ltd.,Shenyang Liaoning 110002,China)
出处
《信息与电脑》
2017年第7期152-154,共3页
Information & Computer
关键词
人工智能
模式识别
支持向量机
入侵
报警
artificial intelligence
pattern recognition
support vector machines
intrusion
alarm
