Machine Learning System For Indolence Perception

Abstract

A person is more prone to nod off while driving, which could cause a traffic accident if they don't receive enough sleep or rest. This leads to a variety of unpleasant scenarios, which we refer to as driver drowsiness. Numerous people are injured or killed in traffic accidents every day throughout the world. According to studies, drivers operating a vehicle when extremely fatigued account for over one-fourth of all fatal highway collisions, suggesting that driver fatigue is a bigger contributor to collisions than drunk driving. This study's main goal is to recognize driver tiredness and decide the best course of action. There are many methods, and they all depend on how the driver is driving or how the automobile is moving. The alert system is one of the physiological strategies utilized to keep the driver attentive and distracted from tiredness. Many strategies are used to deal with expensive sensors and a lot of data. As an outcome, the real-time indolence perception system created in this research has a good method and an acceptable level of accuracy. This prototype system records and captures the driver's facial expressions using a webcam. Each movement in each frame is recognized using a variety of image processing algorithms. Using landmarks, several aspects are calculated, compared, and detected. The outcome is then provided in accordance with the calculated outcome. The alarm system is activated in accordance with comparisons to the current levels.