摘要
为能够准确识别出驾驶员的愤怒情绪,降低道路交通安全隐患,以驾驶员愤怒情绪为研究对象,利用车辆行驶数据建立数据集,构建PCA与改进ANN组合辨识模型,将数据集输入到组合辨识模型中,进行驾驶员愤怒情绪识别,并与单独的改进ANN辨识模型进行对比.结果表明,PCA与改进ANN组合辨识模型的正确率为94.67%,F1-Score为0.8552.相比较于单独的改进ANN辨识模型而言,该模型具有更高的正确率和F1-Score.通过模型的高识别率能够在一定程度上减小甚至避免愤怒情绪所带来的安全隐患问题,为辅助安全驾驶提供了理论依据,进一步提高人机共驾的安全性.
In order to accurately identify the driver’s anger and reduce road traffic safety hazards,this article takes the driver’s anger as the research object,uses vehicle driving data to build a data set,builds a PCA and improved ANN combination identification model,and inputs the data set to the combination identification model.The driver’s anger emotion is recognized and compared with the improved ANN recognition model.The results show that the accuracy of the combined identification model of PCA and improved ANN is 94.67%,and the Fl-Score is 0.8552.Compared with the improved ANN identification model,the model has higher accuracy and Fl-Score.The high recognition rate of the model can reduce or even avoid the potential safety hazards caused by anger to a certain extent,and provide a theoretical basis for assisted safe driving,and further improve the safety of man-machine co-driving.
作者
于祥阁
张敬磊
王云
盖姣云
孙龙祥
YU Xiang-ge;ZHANG Jing-lei;WANG Yun;GAI Jiao-yun;SUN Long-xiang(School of Transportation and Vehicle Engineering,Shangdong University of Technology,Zibo 255000,China)
出处
《数学的实践与认识》
2021年第13期29-40,共12页
Mathematics in Practice and Theory
基金
国家自然科学基金(61573009)
山东省自然科学基金(ZR2017LF015)。
关键词
交通安全
ANN辨识模型
正确率
辅助安全驾驶
人机共驾
traffic safety
ANN identification model
accuracy
assist safe driving
manmachine co-driving
