基于HMM的驾驶员疲劳识别在智能汽车空间的应用  

APPLYING HMM-BASED DRIVER FATIGUE RECOGNITION IN SMART VEHICLE SPACE

作　　者：郁伟炜[1] 吴卿[1] 

机构地区：[1]杭州电子科技大学计算机学院,浙江杭州310018

出　　处：《计算机应用与软件》2011年第10期43-46,共4页Computer Applications and Software

基　　金：国家自然科学基金项目(60703088)

摘　　要：智能汽车空间作为普适计算一种具体而集中的表现,对此提出一个基于隐马尔科夫模型(HMM)的驾驶员疲劳识别应用。选取PERCLOS特征变量作为测评驾驶员疲劳的低层上下文,通过大量样本数据的训练,建立HMM,用Viterbi算法从观察序列中识别出最有可能的驾驶员隐藏状态,提醒驾驶员以确保安全的驾驶行为。最后通过在模拟实验环境中的案例验证了该方法的有效性。Smart vehicle space is a specific and focused performance of pervasive computing;this paper presents an application about driver fatigue recognition based on hidden Markov model（HMM）.The authors select PERCLOS feature variable as a low-level context of driver fatigue evaluation,and establish the HMM through a large number of sample data training.Then they identify the most likely driver＇s hidden state from the observation sequence using Viterbi algorithm,and remind drivers to ensure their safe driving behaviour.Finally,a case study in simulation environment confirmed the validity of the scheme.

关 键 词：智能汽车空间 隐马尔科夫模型 上下文推理 驾驶员疲劳 PERCLOS 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]