摘要
针对现有车载安全电子产品功能单一、对风险的预判存在局限性的问题,提出一种智能行车安全预警系统设计方案.本方案采用机器视觉技术检测车内驾驶员疲劳驾驶和抽烟等危险驾驶行为,识别交通信号、交通标志和障碍物等车外路况信息,以及采集车辆的行驶状态,并综合上述信息预判出发生交通事故的可能性并做出相应的预警提示.实验结果表明,本方案可行性高,复合判断方法比单一方法识别检测的准确率高.
A set of intelligent driving safety warning system is proposed for improving the singularity of the existing vehicle-mounted safety electronic product and the low efficiency of traffic accident prevention. It uses machine vision to realize detection of driver＇s fatigue and unsafe driving behavior such as smoking. At the same time, it can identify traffic signals, traffic patterns, obstacles and off-road conditions which record the driving status of vehicles. Based on the above information, it can quickly determine the possibility of the accident and make the corresponding warning prompt. The results show that the scheme is feasible and of higher accuracy than the single method.
作者
林虹
吴良峰
LIN Hong;WU Liang-feng(Department of Information Engineering,Yango University,Fuzhou,Fujian 350015,China;Fuzhou Roekehips Electronics CO.,Ltd.,Fuzhou,Fujian 350003,China)
出处
《宁德师范学院学报（自然科学版）》
2018年第3期324-329,共6页
Journal of Ningde Normal University(Natural Science)
基金
福建省教育厅中青年教师教育科研项目(JA15628)
