Paper-03 | Reg. No.:20140903|DOI:V3I2P03
Vision for Driver Assistance: A Real Time Drowsiness Detection Based on Template Matching
Priyanaka P. Sawant , Dr.M.Z.Shaikh,Bharati vidyapeeth college of engineering, Kharghar, Navi Mumbai.
This paper presents the template matching technique to detect driver fatigue or drowsiness. In modern days, as technology kept going on higher echelons, the number of automobiles on the road increased and the accidents too. The percentage of accidents caused by human error is very high. We tried to ease and make safe drivers life by having certain tools like navigation systems, speed limit alarms etc. However the main cause of the accident was the fatigue and drowsiness. There cannot be anything dangerous than sleepy driver on steering wheel – for himself, passengers and people outside. Understanding the criticality of the problem, there was the need for the system that would detect the drowsiness or fatigue. There have been much work and research to build such system. One of the key strategy for drowsiness detection technique was using Eye tracking. An eye is a very critical component to reflect the current state of the driver. Driver’s attention level, concentration and fatigue can be understood by monitoring driver’s eye. One more reason why monitoring eye will be useful is that, the attention of driver towards driving. If the completely awaken driver is busy in the conversation with passengers or involved in any other distraction which makes him lose the track of road leads to accidents. During eye tracking, the first step is to identify the face, then identify the eyes. Once the system has fixed its focus on them, the eyes are tracked at a real time using the pre-obtained template. This result is compared with the vision based data obtained and compared with the threshold for the fatigue. When the one is reached, the alarm can be triggered to mitigate the further risk.