基于DSP多特征检测疲劳驾驶监测系统研究  

ON FATIGUE DRIVING MONITORING SYSTEM BASED ON DSP MULTI-CHARACTERISTICS DETECTION

作　　者：张海水[1] 詹永照[1] 刘志强[2] 

机构地区：[1]江苏大学计算机科学与通信工程学院,江苏镇江212013 [2]江苏大学汽车与交通工程学院,江苏镇江212013

出　　处：《计算机应用与软件》2009年第9期61-63,共3页Computer Applications and Software

基　　金：江苏省高新技术项目(BG2005028);国家科技支撑计划课题(2007BAK35B02)

摘　　要：针对疲劳驾驶监测的要求,提出基于DSP红外条件下对眼睛、嘴巴、头部同时检测进行疲劳监测的方法。该方法首先利用人眼红外光源下的红外效应来进行图像差分,以此获取瞳孔来定位眼睛区域;然后用基于个体的模板匹配方法进行特征提取和动态模板匹配方法进行原始迭代跟踪;最后用计算出来的眼睛的闭合程度和是否打哈欠、点头来判断疲劳程度。实验证明该方法综合了驾驶员的三个个体特征。准确率较高、速度较快,可以满足非接触式、全天候、实时监测的要求。Aiming at the demands of fatigue driving monitoring, a DSP based driver fatigue detection method which detects eyes, mouth and head simultaneously is proposed. The method first locates eyes region by the pupils which are obtained from the difference processing using in- frared effect under infrared light source. Then the individual-based template matching is used to extract the features and the dynamic template matching is used to process primitive iterative trace. Finally, the degree of eyes closing calculated together with yawning and nodding are utilized to judge the degree of the fatigue. Experiments show that this approach combines three individual characterises of a driver. It has higher accuracy and speed, and can satisfy the demands of non-contact, all illumination condition and real-time monitoring.

关 键 词：疲劳监测 DSP 模板匹配 IR 实时监测 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术] TD76[自动化与计算机技术—计算机科学与技术]