Driver Drowsiness Detection and Alert System

Road accidents caused by driver fatigue and inattention are on the rise, with drowsy driving incidents occurring more frequently. This research is dedicated to strengthening efforts in detecting driver drowsiness under real driving conditions, aiming to minimize the number of traffic accidents. By reviewing prior studies on drowsiness detection systems, several methods have been explored to effectively identify signs of driver fatigue and inattentiveness. The objective of this project is to create an interface capable of automatically detecting drowsiness in drivers through live images captured via a webcam. Advanced machine learning algorithms will be employed to process these images and determine whether the driver is experiencing fatigue. When drowsiness is detected, a buzzer alarm will be triggered, progressively increasing in volume. If the driver fails to respond, a warning message in the form of a text and an email will be dispatched to their family members, notifying them of the situation. The primary goal of this system is to identify if the driver is in a sleeping or drowsy state. The methodology involves real-time image acquisition from a webcam, followed by facial and eye feature extraction using the dlib library, ensuring accurate drowsiness detection and contributing to improved road safety.