Enhanced Drowsiness Detection System with Haar Cascade and Dlib using Eye Aspect Ratio

Drowsiness can have serious consequences, such as decreased performance and an increased risk of accidents. Existing systems for drowsiness detection rely on physical indicators, which can be unreliable or inaccurate. This work proposes a novel approach to drowsiness detection using a combination of the Haar cascade algorithm, Dlib, and EAR calculations. Our method can accurately detect drowsiness by analyzing the changes in eye movement, a reliable indicator of drowsiness. The proposed method is effective and efficient, making it a promising solution for real-time drowsiness detection applications. Various methods in the literature for detecting drowsiness, such as electroencephalography (EEG), heart rate variability (HRV), and facial expressions. However, these methods have cost, invasiveness, and reliability limitations. Our approach is effective but also non-invasive and cost-effective, making it a promising solution for improving safety and performance in various contexts.