Paper Title
An improved driver monitoring system for Accident prevention

Accidents are not only caused by poor technical conditions of the vehicles, but also by tired, indisposed, or bad state-of-minded drivers. Driver inattentiveness has been identified as one of the principal causes of accidents on road. This paper presents an image based, real time driver attention monitoring system to detect early symptoms of drowsiness. This study presents an approach to detect driver’s drowsiness in advance by applying two distinct methods in computer vision and image processing. Using Haar-Classifier Algorithm (HCA), face is detected in an image and then using Hough Transformation, accurate localization of eye is detected and regionalized from the face image. Once eye is detected, the Blink detection can be calculated by using a motion detector based on threshold frame difference inside the tracked regions of interest. Again using eye closure rating information on this "inattentive" eye category, inattentiveness is quantified and above a certain threshold value an alarm sound is generated to indicate driver inattentiveness.