Drowsiness Detection Using Adaboost Method and Haar Cascade Classifier to Improve Safety of Drivers

Drivers who need a nap are to blame for one-fourth of all major highway accidents, suggesting that tiredness is a bigger factor in crashes than drunk driving. Drowsiness detection systems operate in real-time, continuously monitoring the individual's state and providing timely alerts when drowsiness is detected. This allows for proactive intervention before drowsiness leads to accidents or errors. The primary objective of this research endeavor is to develop a non-intrusive system capable of detecting human fatigue and providing an immediate warning. The proposed solution employs the employment of a camera to track the driver's vision to recognize signs of driver fatigue enough in time to prevent the driver from sleeping. This will be important for anticipating driver fatigue and presenting caution information in the manner of alerts and popups. In addition, instead of being deleted by itself, the notification will be deleted manually. If the driver is fatigued, they may react improperly to the conversation. If all three input parameters display an increased risk of fatigue at the same moment, a time domain graph is produced. The warning message is delivered through text message. This directly indicates sleepiness or exhaustion, which is then utilized as a record of driving performance.