Driver Fatigue Detection Based On Facial Features

In recent years,with the increase of car ownership,the frequency of traffic accidents has remained high,resulting in a large number of casualties and huge economic losses.In many accidents,fatigue driving accidents account for a large proportion,so driver fatigue detection technology has attracted the attention of scholars at home and abroad.Combining the eye and mouth characteristics,fatigue analysis is carried out.Compared with previous studies,the problem of detection accuracy is reasonably solved.It has the advantages of non-contact,real-time and so on.It is of great significance to ensure driver’s personal safety.Firstly,the driver’s image is preprocessed,i.e.image enhancement,filtering and light compensation,to improve the quality of the image,reduce noise interference and make the face receive light evenly.Secondly,face location is carried out by AdaBoost algorithm based on Haar feature.By constructing weak classifier,the best Haar feature is selected to construct strong classifier,which is cascaded as the best strong classifier,and then face detection is carried out.Then the recognition and location of eyes and mouth are carried out,and the eye region is located based on integral projection method.The eye opening and closing state is analyzed by calculating the percentage of black pixels of eyelids and pupils to the total pixels of eye region.The mouth region is located based on the improved "three chambers and five eyes" method,and then the height-width ratio of mouth is calculated by integral projection method to judge the mouth.The state of opening and closing.Finally,according to PERCLOS criterion and blink frequency,eye-based fatigue analysis is carried out,mouth-based fatigue analysis is carried out according to yawning,and driver’s fatigue analysis is carried out combining eyes and mouth.Through the experimental results,it can be seen that the fatigue state analysis combined with the eye and mouth features greatly improves the detection accuracy compared with the single feature analysis.It can detect the driver in real time and accurately,which is of great significance for ensuring the driver’s personal safety.