摘要
为提升城市道路交通安全管理效率,减少交通事故发生率,设计一种基于地理信息系统(geographic information system,GIS)和遗传算法优化神经网络的城市道路交通安全自动预警方法。首先利用GIS软件MapInfo遍历目标道路网络中的每一道路元素,实施事故多发点自动搜索,筛选并标注出事故频发点,作为自动预警点位;采集其交通流量数据、道路状况数据、天气条件数据、交通管控措施数据、社会经济数据、时间因素数据,进行缺失值处理与规划处理;结合遗传算法优化的神经网络实施道路交通事故预警。测试该方法的预警效果及误报次数,对比基于贝叶斯网络预警方法(方法A)、基于机器视觉与信息共享的预警方法(方法B),实验测试结果表明,该方法的误报次数明显少于方法A和方法B,预警效果显著,能够有效地预警潜在的交通事故,减少误报率,提高了预警系统的效率和准确性,为城市交通安全管理提供了新的思路。
To enhance the efficiency of urban road traffic safety management and reduce traffic accident rates,this study designs an automatic early warning method based on geographic information system(GIS)and a neural network optimized by a genetic algorithm.First,GIS software MapInfo is used to traverse each road element within the target road network,automatically identify accident-prone spots,screen and mark frequent accident points,and designate them as automatic warning points.Data including traffic flow,road conditions,weather conditions,traffic control measures,socio-economic factors,and time-related factors are collected,followed by missing value processing and planning processing.Road traffic accident warnings are then implemented using a neural network optimized by a genetic algorithm.The early warning effectiveness and false alarm rate of the proposed method are tested and compared against a Bayesian network-based warning method(Method A)and a machine vision and information sharing-based warning method(Method B).Experimental results demonstrate that the number of false alarms of the proposed method is significantly lower than that of Method A and Method B.The proposed method shows remarkable warning performance,can effectively warn of potential traffic accidents,reduces the false alarm rate,improves the efficiency and accuracy of the early warning system,and provides a new approach for urban traffic safety management.
作者
何增镇
HE Zengzhen(Digital Intelligence Engineering Division,Guangxi Transportation Science And Technology Group Co.,Ltd.,Nanning 530007,Guangxi,China)
出处
《自动化技术与应用》
2026年第4期173-176,185,共5页
Techniques of Automation and Applications
基金
广西壮族自治区重点研发计划项目(桂科AB23026155)
交通运输部交通运输行业重点科技项目清单(2022-ZD4-052)。
