摘要
本文针对船舶驾驶场景,提出了一种基于随机森林和BP神经网络的船舶驾驶员疲劳检测算法,及时检测船舶驾驶员的疲劳驾驶行为,对安全航行有重要意义。首先,使用随机森林算法对特征数据按照场景分类;其次,根据场景将数据分发到不同的BP神经网络模型进行预测;最后,使用船舶仿真模拟器进行实验验证。结果显示,算法识别准确率为0.82,召回率为0.65,精确率为0.69,说明本文提出的算法对船舶驾驶员疲劳驾驶行为检测有一定实用价值,且与生理数据监测方法相比,简单方便、成本更低,对驾驶员无干扰。
This paper proposes a ship pilot fatigue detection algorithm based on random forest and BP Neural Network for maritime navigation scenarios,which plays a significant role in ensuring safe navigation by timely detecting fatigue driving behaviors of ship pilots.Initially,the Random Forest algorithm is used to classify feature data according to the scenario.Subsequently,the data is distributed to different BP Neural Network models for prediction based on the scenario.Finally,a ship simulator is used for experimental validation.The results show that the algorithm has an accuracy of 0.82,a recall rate of 0.65,and a precision of 0.69.This demonstrates that the algorithm proposed in this paper has practical value in detecting fatigue driving behaviors of ship pilots.Compared to physiological data monitoring methods,it is simpler,more convenient,lower in cost,and non-intrusive to the pilots.
作者
张威特
李俊松
刘雁飞
ZHANG Weite;LI Junsong;LIU Yanfei(School of Computer Science and Technology,Zhejiang Sci-Tech University,Hangzhou 310018,China)
出处
《智能计算机与应用》
2024年第2期140-143,共4页
Intelligent Computer and Applications
基金
浙江省自然科学基金(LY12C09005)。