Research Of Driver's Drowsy Warning System Based On Face Character Recognition

In Modern times, while the expressway is developing excessively, drowsy driving as one of the most severe factors caused traffic accident. The bad accident not only brings serious problem and great effect to society, but also causes hell and gone loss of property and precious lives. So in such bad environment, it is necessary and meaningful to study the detection of driver's drowsy status and auxiliary drive device. On the base of face detection, This paper studies the application of the driver's drowsy status by the method of PCA(Primary Component Analysis). Through analyzing the face video, computer makes the right disposal according to driver's drowsy status judged by the mouse and eyes status.On the base of studying the driver's drowsy status detection,face detection and face recognition in computer vision, the main task this paper has done present below:(1)Using the driver's face skin identity to position the skin region, as study indicate, the face skin distribute relatively centralized in color space. The color information could distinguish between face region and non-face region to some extent. in succession, this paper indicate the region of mouse and eyes by the face framework. Because of the fast computing and simple course, skin information arithmetic meet the demand of real time, at the same time, the face skin identity is highly adaptive to the geometry variation such as face zoom and face circumrotation.(2)In constructing drowsy status recognition system, this thesis studies necessary image pretreatment technique like illumination compensation, two-value image treat. In the application of confirming driver's face region, the thesis studies real time face detection arithmetic by the way of coordinate projection which did some advancement in the aspect of real time. Studying the mouse and eyes distribution in face region, it makes much more progress in real time.(3)This thesis attach great importance to the PCA recognition method in identifying mouse status and eyes status, achieving good result. Simplifying all the status of mouse and eyes to three forms: fully open,half open,close, and then detect blink and yawn according to their relationship, confirming the blink sustained time and blink frequency and other parameters related, confirming the yawn sustained time and yawn frequency and other parameters related .In this thesis's driver drowsy admonition system, we get local face image character in detected face image, comparing the face character with which stored in driver's image database, match the right one if existed and complete drowsy status recognition finally. As well as constructing the driver drowsy admonition system, this paper did some improvement in automated face detection and automated drowsy recognition, which adapt to this system, meet the rightness and real time demand.