This research paper explores the application of computer vision techniques for detecting symptoms of drowsiness, primarily in drivers, to reduce road accidents. The demanding work schedules, especially those of commercial truck drivers, cause them to drift off while driving, leading to dangerous scenarios that could jeopardize many lives. In an attempt to solve this problem, a convolutional neural network (CNN) model was employed. The model was trained using an augmented dataset comprising 350,000 images, aiming to detect symptoms of drowsiness, such as full and partial eye closure, yawning, and drooping of the head. Real-time detection was enabled by interfacing the model with a live video feed from a camera, processing approximately one frame per second, due to hardware constraints during testing. To enhance the model’s robustness, a pretrained HAAR cascade face detector from OpenCV was integrated. This ensured the extraction of only the subject’s face, making the model independent of the camera’s position during predictions. The model achieved an accuracy of 90.13%. The outcomes of this research highlight the effectiveness of the proposed approach in accurately and efficiently detecting symptoms of drowsiness. The application of computer vision and the utilization of CNN algorithms showcase promising prospects for the development of practical systems to enhance road safety. These systems can monitor driver fatigue in real time, thereby contributing to overall improvements in road safety.