Pupil localizing in video images the first step toward eye monitoring

In this paper we propose a method for pupil localizing in infrared video images. Pupil localizing is the first step toward eye monitoring. The proposed method is based on three phases. At first the pupil is segmented Pupil segmentation is based on thresholding. The threshold value is estimated on the basis of a genetic algorithm using Otsu thresholding technique. Following, the pupil is modeled. We assumed the pupil as a circle therefore the modeling process is based on circular Hough transform Finally, the pupil tracking is performed. Center and radius of pupil are the parameters used for tracking. Applying the Kalman filter, pupil movements are estimated in video sequences. Experimental results show promising performance.