QQ客服 官方微博 反馈留言

题名	基于人脸特征识别的疲劳驾驶预警系统; Driver Fatigue Detection System Based on the Feature of Face
作者	朱清坤
答辩日期	2012 ; 2012
导师	副教授
关键词	驾驶疲劳检测 Fatigue Driving Monitoring PERCLOS PERCLOS 多特征 Multi_features LBP LBP
英文摘要	随着社会经济的发展，机动车辆与日俱增，道路交通事故频发，死亡人数一直居高不下。疲劳驾驶是引发频繁的交通事故的主要因素之一。因此，如何有效的监测和防止驾驶员疲劳驾驶，对于避免交通事故，提高交通安全性有着重要意义。基于机器视觉的疲劳检测，在实时性、非接触性以及全天候等方面有更大的优势，所以它成为疲劳检测的一个主要方法。 本文在研究了国内外相关研究的基础上，设计了一套改进的疲劳监测系统，该系统体积小、便于安装，可以满足全天候、实时监测的要求。本文主要研究内容如下： (1) 研究总结现有的人脸检测方法，选用AdaBoost算法作为人脸和人眼检测定位方法。实验表明，本文的方法增强了分类器的分类能力， 提高了系统的正确检测率，降低了错误报警率。并且提出了一种新的采用局部特征改进算法进行眼睛特征的提取。大量的实验表明这种方法能够准确地定位眼睛和检测眼睛的疲劳状态。 (2) 针对夜间光照不足的情况，提出了采用红外下差分图像与图像处理相结合的瞳孔定位算法。在不同的红外光下获取的奇、偶帧图像进行差分，对差分图像进行平滑滤波，然后选择合适的阈值二值化分割图像，最后提取眼睛瞳孔。该方法定位眼睛快速、准确。 (3) 提出了基于LBP的嘴唇定位和嘴部分析的打哈欠疲劳检测方法。实验的结果表明这种方法检测准确率较高，可用于对驾驶员打哈欠的疲劳检测。 (4) 提出了PERCLOS值大小、哈欠的次数及其持续时间长短两个参数进行疲劳状态判断。该方法综合多个特征对疲劳状态进行判断，提高了判断的准确度。; With the development of society and economy, the frequent traffic accidents and high mortality due to the increasing number of vehicles have become a serious problem for society. Driver fatigue is one of the main causes of frequent traffic accidents. In this case, monitoring and preventing drivers from driving in fatigue effectively has a very important meaning at avoiding traffic accidents and enhancing traffic safety. Fatigue detection based on machine vision has great advantages in real time, non-contact and all-weather monitoring. It has become a main method of the fatigue detection. Based on the domestic and foreign research, this paper propose an improvement fatigue detection system which has the advantages of small size and being installed easily, meeting the requirements of a all-weather working and real time monitoring. In this paper, main research methods are as follows: (1) According to the existing face detection methods, a face detection algorithm based on AdaBoost is used to detect the face and eyes. The experimental results show that the method improved the performance of classifiers, accurate detection rates and decrease the classifiers’ false alarm rates. A new algorithm which is use the local feature on the feature of eye extraction is designed. A great deal of experiments are done, which show that human’s eyes and eye fatigue are accurately monitored with this method. (2) In view of low-level lighting at night, we propose an algorithm which combine infrared radiation difference picture with image processing.According to the algorithm, the difference is first carried out on the odd and even frame of infrared radiation image. Then smoothing filter and binarization with a suitable threshold are conducted on the difference image. Finally, pupil can be localed exactly. This algorithm has the advantages of fast and accurate for eyes location. (3) Yawning detection algorithm of lip locating and mouth shape analysis based on LBP is presented. The experimental results show that detection method has high performance. This method could be used for yawning detection to determine driver fatigue state. (4) A fatigue status monitor method is proposed which bases on two parameters including PERCLOS value, the number and the duration of yawning. The method that combine multi-features parameters for driver fatigue can improve detection accuracy significantly.; 学位：工程硕士; 院系专业：信息科学与技术学院通信工程系_电子与通信工程; 学号：23320091152790
语种	zh_CN
出处	http://210.34.4.13:8080/lunwen/detail.asp?serial=36009
内容类型	学位论文
源URL	[http://dspace.xmu.edu.cn/handle/2288/51455]
专题	信息技术－学位论文
推荐引用方式 GB/T 7714	朱清坤. 基于人脸特征识别的疲劳驾驶预警系统, Driver Fatigue Detection System Based on the Feature of Face[D]. 2012, 2012.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。