FATIGUE STATE DETECTION METHOD BASED ON FACIAL FEATURE POINT POSITIONING

摘要

Disclosed is a fatigue detection method based on facial feature point positioning. A user aims a camera at a face, and during detection after an initialization stage, image of each frame inputted by a monitoring video is detected using a face detection and facial feature point positioning algorithm to obtain facial feature points. An eye-related area is determined according to the facial feature points of the user, and whether the user has closed eyes is determined according to a degree of change in a grayscale value of the eye-related area, and an alert concerning a fatigue level of the user is issued according to an eye closure condition in an adjacent frame. The present invention is simple to use, has a small amount of calculation, is accurate in determination, may be applied to a real-time environment, and effectively reminds the user to rest, thus ensuring life and property security of the user.