基于云模型的驾驶疲劳检测方法
|
摘要
疲劳驾驶是导致交通事故的重要原因。因此实时监测驾驶员的疲劳状态具有重要的现实意义。文中针对人的眼动行为存在随机性及模糊性的特点,采用不确定性的云概念,建立基于Perclos和眨眼时间均值的双条件单规则发生器,然后在此基础上构造基于双条件单规则发生器的疲劳检测模型。最后通过在模拟实验平台上采集驾驶员的眼动数据,将眼动特征数据输入基于双条件单规则发生器的疲劳检测模型进行疲劳识别判断。实验结果表明:该模型的平均识别率达到80.2%。在同组实验数据下,使用云模型方法比采用K最近邻(KNN)及支持向量机分类算法的检测率都要高。
Fatigue driving is one of the causes of traffic accidents.Therefore,it is of great practical significance to monitor the driver's fatigue state in real time.In view of the randomness and fuzziness of human eye movement behavior,a double-condition single rule generator based on Perclos and blink time mean is established by using the concept of uncertain cloud.And then a fatigue detection model based on double-condition single rule generator is constructed.Finally,by collecting the driver's eye movement data on the simulation experimental platform,the eye movement characteristic data is put into the fatigue detection model based on the double-condition single rule generator to identify and judge the fatigue.The experimental result shows that the average recognition rate of the model reaches 80.2%.In the same group of experimental data,the detection rate using this method is higher than using KNN(K-nearest neighbor)and SVM(support vector machine)classification algorithm.
Fatigue driving is one of the causes of traffic accidents.Therefore,it is of great practical significance to monitor the driver's fatigue state in real time.In view of the randomness and fuzziness of human eye movement behavior,a double-condition single rule generator based on Perclos and blink time mean is established by using the concept of uncertain cloud.And then a fatigue detection model based on double-condition single rule generator is constructed.Finally,by collecting the driver's eye movement data on the simulation experimental platform,the eye movement characteristic data is put into the fatigue detection model based on the double-condition single rule generator to identify and judge the fatigue.The experimental result shows that the average recognition rate of the model reaches 80.2%.In the same group of experimental data,the detection rate using this method is higher than using KNN(K-nearest neighbor)and SVM(support vector machine)classification algorithm.
引文
[1]肖赛,雷叶维.驾驶疲劳致因及监测研究进展[J].交通科技与经济,2017,19(4):14-19.
[2]牛清宁,周志强,金立生,等.基于眼动特征的疲劳驾驶检测方法[J].哈尔滨工程大学学报,2015,36(3):394-398.
[3]潘晓东,李君羡.基于眼部行为的驾驶疲劳监测方法[J].同济大学学报(自然科学版),2011,39(2):231-235.
[4]张驰,马广露,朱国华.驾驶疲劳检测技术概述[J].交通工程,2018,18(1):18-22.
[5]郭敏,李宁.一种基于云模型的负荷频率控制方法[J].电子设计工程,2018,26(6):157-160.
[6]李德毅,杜鹢.不确定性人工智能[M].北京:国防工业出版社,2005:44-46.
[7]周红波,李照顺,谢佑波.基于云模型的系统综合效能评估方法[J].火力与指挥控制,2017,42(8):61-63.
[8]费金龙,王禹,王天鹏,等.基于云模型的网络异常流量检测[J].计算机工程,2017,43(1):178-182.
[9]高洪波,张新钰,张天雷,等.基于云模型的智能驾驶车辆变粒度测评研究[J].电子学报,2016,44(2):365-373.
[10]金菊良,宋占智,蒋尚明,等.基于云模型的淮北平原参考作物蒸散量时空分布[J].南水北调与水利科技,2017,15(1):25-32.
[11]王玲俊,王英.基于云模型的装备制造业产业链风险评价[J].技术经济,2016,35(2):80-87.
[12]王丹丹,李美花,崔家佐.基于逆向条件云发生器的评价模型研究[J].信息与电脑(理论版),2017(23):49-50.
[13]江浩斌,吴狄,沈峥楠,等.基于云发生器的自动泊车路径跟踪控制研究[J].机械设计,2016(9):61-65.
[14]张明媛,高盼,袁永博.云模型理论在区域旱涝评价中的应用[J].模糊系统与数学,2016(1):174-181.
[15]陈昊,李兵,刘常昱.一种无确定度的逆向云算法[J].小型微型计算机系统,2015,36(3):544-549.
[2]牛清宁,周志强,金立生,等.基于眼动特征的疲劳驾驶检测方法[J].哈尔滨工程大学学报,2015,36(3):394-398.
[3]潘晓东,李君羡.基于眼部行为的驾驶疲劳监测方法[J].同济大学学报(自然科学版),2011,39(2):231-235.
[4]张驰,马广露,朱国华.驾驶疲劳检测技术概述[J].交通工程,2018,18(1):18-22.
[5]郭敏,李宁.一种基于云模型的负荷频率控制方法[J].电子设计工程,2018,26(6):157-160.
[6]李德毅,杜鹢.不确定性人工智能[M].北京:国防工业出版社,2005:44-46.
[7]周红波,李照顺,谢佑波.基于云模型的系统综合效能评估方法[J].火力与指挥控制,2017,42(8):61-63.
[8]费金龙,王禹,王天鹏,等.基于云模型的网络异常流量检测[J].计算机工程,2017,43(1):178-182.
[9]高洪波,张新钰,张天雷,等.基于云模型的智能驾驶车辆变粒度测评研究[J].电子学报,2016,44(2):365-373.
[10]金菊良,宋占智,蒋尚明,等.基于云模型的淮北平原参考作物蒸散量时空分布[J].南水北调与水利科技,2017,15(1):25-32.
[11]王玲俊,王英.基于云模型的装备制造业产业链风险评价[J].技术经济,2016,35(2):80-87.
[12]王丹丹,李美花,崔家佐.基于逆向条件云发生器的评价模型研究[J].信息与电脑(理论版),2017(23):49-50.
[13]江浩斌,吴狄,沈峥楠,等.基于云发生器的自动泊车路径跟踪控制研究[J].机械设计,2016(9):61-65.
[14]张明媛,高盼,袁永博.云模型理论在区域旱涝评价中的应用[J].模糊系统与数学,2016(1):174-181.
[15]陈昊,李兵,刘常昱.一种无确定度的逆向云算法[J].小型微型计算机系统,2015,36(3):544-549.