Real Time Slumbering Detection Using DLIB
Harsh Agrawal, Pallavit Saxena, Shiva Tiwari, Sandhya Awasthi, Kadambri Agarwal · 2024
Worldwide, driving when fatigued is still a major factor in traffic accidents. Our solution to this problem is a complete slumbering detection system that uses real-time monitoring, machine learning algorithms, and computer vision techniques to identify indicators of driver sleepiness. To recognize signs like eye closure, yawning, and head movement, our system makes use of many libraries, including dlib for facial landmark identification and OpenCV for face detection. Furthermore, the driver's past history of sleepiness is also stored in the database. To prevent tiredness the next time, the database viewer might take proactive steps to enhance his or her sleep. The likelihood of feeling drowsy increases significantly at night compared to the day. It should thus be possible to have a system that operates differently during the day and at night. We may tighten our system at night so that we are warned if someone yawns once or twice, and we can be alerted if someone yawns three or four times during the day. Head movement follows similar rules: during the day, we may ignore some head movement, while at night, we can pay close attention to even the tiniest movement.