Pupil track generation method based on human eye video

The embodiment of the invention discloses a pupil trajectory generation method based on a human eye video, which comprises the following steps of: 1, performing coarse filtering on each frame of image, performing primary morphological filtering, and eliminating white spots formed at pupils under the irradiation of infrared light; 2, calculating a minimum gray average value of each frame of image, judging whether the frame is an effective frame according to the minimum gray average value, and if so, taking a threshold value; step 3, carrying out binarization processing; 4, performing morphological filtering for the second time, removing eyebrows and small skin pigment areas, and filling the pupil areas after binaryzation; step 5, performing edge detection by using a Laplacian operator to obtain the contour of the pupil; and step 6, calculating a subpupil pole. According to the method, the Agt can be effectively treated; the frame is S * 10%, and the situation that eyes are closed or pupils are blurred basicall