ENHANCING ROAD SAFETY WITH REAL TIME DRIVER DROWSINESS DETECTION

There has been greater concern about the driver’s drowsiness on road safety. According to the survey of the National Highway Traffic Safety Administration (NHTSA) a greater percentage of fatalities, injuries and even deaths every year is because of drowsy driving. So, there is an immediate necessity to implement a system which detects the drowsiness of the driver and alerts the driver. These systems, which rely on visual behaviour analysis, hold the potential to significantly decrease accidents by providing timely alerts when drivers exhibit signs of drowsiness. These systems make use of cameras and computer vision algorithms, such as the Haar cascade classifier and CNN. These systems scrutinise facial features, eye movements, and other indicators to assess levels of alertness and identify signs of drowsiness. The cameras that are integrated continuously capture facial expressions, enabling the evaluation of eyelid closure for the Eye Aspect Ratio (EAR) and Mouth aspect ratio (MAR) across frames. If predefined thresholds for EAR values are surpassed, an alert system triggers, notifying both the driver and passengers.