Fatigue Driving Detection Based On Facial Analysis

Studies shows that driver fatigue is one of the main causes of traffic accidents.Therefore,people pay more attention to driving fatigue detection.A face analysis method is adopted in this thesis to detect driving fatigue,which is of great significance in prevent driver fatigue and improve driving safety.This thesis mainly includes the following contents.The blink detection method based on eye aspect ratio ignores the gray information of the eye region,and the error rate is high.Therefore,a blink detection algorithm based on gray information is proposed,the core of the algorithm is to use the gray information of the eyeball region to get a threshold and then binarize the eye region.Eyeball region is separated after pre-processing,and the blink detection is carried out by the proportion of black pixel values in the eye region.Compared to the single feature method,the method of using two features for blink detection is more accurate.In this thesis,an direct and simple algorithm for detecting the yawning by mouth aspect ratio mouth is proposed.Based on the improved blink and yawn detection algorithm,various fatigue features of eyes,mouth and head are used for driving fatigue detection,which is more accurate compared with extracting single driving fatigue feature.In this thesis,a driving duration detection method based on face verification technology is proposed.By continuously monitoring the driver’s identity and analyzing various situations in the driving process,a reasonable method for detecting driving duration is proposed.Finally,the fatigue driving detection platform is designed to verify the algorithm proposed in this thesis.The results show that the algorithm is accurate and robust.