IoT Based Driver Drowsiness Detection Using Convolutional Neural Network

Drowsiness while driving is a core threat to global safety on roads, resulting in significant traffic accidents and fatalities. Efficient strategies to combat drowsy driving are an emerging need. The growing volume of vehicle traffic demands machine learning based technological advancements and scalable solutions to prevent road accidents. This paper discusses an IoTbased hybrid driver drowsiness detection system using machine learning algorithms. Cameras and sensors are used to capture the real time video of the driver's face and eyes, steering angle, vehicle speed etc. Additional sensors are used to monitor the heart rate, skin temperature etc. The YOLO model is used to extract features like eye blinking rate and yawning detection and to localize facial features like eyes, mouth, and head position with high precision. The identified regions from YOLO are passed to a CNN for further analysis and drowsiness detection. The proposed model is compared with the existing system and it outperforms it with the highest accuracy of 88.6%.