Driver’s Drowsiness Detection System Using Machine Learning

The structure focuses on the eyes to prevent accidents. In the proposed system, a video is first recorded by a camera. The camera is placed in front of the driver to better capture the driver's face. To create a 2D image, frames of the video are extracted. Faces are detected in images using histogram of oriented gradients (HOG) and linear support vector machine (SVM) for object detection. In this case, SVM (Support Vector Machine) was used to classify the components of the video input. To do this, we first need to develop a bounding box and classification system to recognize pupils initially. Following facial recognition, it marks facial points on the image, such as eye positions. Eye aspect ratio (EAR) and mouth aspect ratio (MAR) are measured using facial landmarks. These features and machine learning approaches are used to estimate driver fatigue. If it detects drowsiness, it sends an alert to the driver to alert them. This system uses the SVM (Support Vector Machine) approach which is achieving an excellent accuracy score of 92.85%. This approach improves model performance and helps to reliably detect driver fatigue and prevent fatal crashes.