Driver Fatigue Countermeasure System

One of the major contributing factors to traffic accidents is driver weariness and drowsiness. Automated driver sleepiness detection is a significant computer vision challenge since accidents involving fatigued drivers have resulted in millions of fatalities in recent years. A real-time driver sleepiness detection system is created to address this issue and is based on Facial Landmark Detection. A significant barrier still exists in accurately identifying driver fatigue and forecasting it in real-time scenarios. In the proposed method, drowsiness detection is done by the fusion of eye state recognition and yawn detection. This study describes a non-intrusive driver fatigue detection system that can identify upcoming driver drowsiness and send out timely warnings that could be helpful. The system relies on visual cue, here mouth and eye to characterize the driver’s alertness level. To detect the mouth and eyes from the full face region, the system uses a Haar cascade classifier function and is followed by extraction of mouth and eyes from images. It also uses image processing techniques to extract the driver’s mouth and eyes to determine their states. The proposed model is built using Python language and Open CV.