An Edge-based Method to Classify Open and Closed Eyes for Monitoring Driver’s Drowsiness
Jaeik Jo, Sung Joo Lee, Youn Joo Lee, Ho Gi Jung, Kang Ryoung Park, Jaihie Kim · 2010
Abstract- Monitoring the drowsiness and inattention of a driver plays an important role in preventing car accidents. Previous researches show that many vehicular accidents in highway are caused by the fatigue due to long-time driving, inattention or dozing of a driver. To resolve these problems, we propose a method to classify open and closed eyes by analyzing horizontal edge of eye images. After initial locations of face and eyes are found by using object detection methods such as Adaboost, an edge-based method is used to extract feature from detected eye image. Then, open and closed eyes are classified by the threshold based on Bayesian rule and analysis of extracted feature. Experimental results show that open and closed eyes are reliably classified in actual vehicular environments. I.