Deep Learning based Drowsiness Detection and Monitoring using Behavioural Approach
P. William, Mohd Sarim Shamim, Ajay Reddy Yeruva, Durgaprasad Gangodkar, Swati Vashisht, Amarendranath Choudhury · 2022 2nd International Conference on Technological Advancements in Computational Sciences (ICTACS) · 2022
Using deep learning and a behavioural approach, this study presents a real-time detection and monitoring system for tired drivers. The objective is to develop and build software that collects real-time driver behaviour while driving and trains it using convolutional neural networks (CNNs) to anticipate the driver's behaviour. An intelligent video-based gadget, a dataset of drowsy drivers, and CNN architecture were used to achieve this goal. MATLAB and a deep learning technology were used to implement the concepts. Tests revealed that the system has a 99.8% accuracy rate for detecting anomalies. A prototype model of the system was created using MATLAB.