Basically, this project is about Driver Drowsiness that is the major cause for traffic accidents. Our solution aims towards alerting drivers on real time and resulting in prevention of accidents. Taking images from mobile camera and feed them into a Deep Learning model which will classify whether the person eyes are ‘open or ‘closed’. Detect the face in the image from ROI and feed it to classifier. Classifier will categorize whether eyes are open or closed. This system works by monitoring the eyes of the driver and sounding an alarm when he/she is drowsy. The priority is on improving the safety of the driver without being obtrusive.The approach adapted will to use edge device’s camera (mobile phone) and feed the stream to neural network, that will classify the driver as fit or unfit for driving in particular scenario. Our goal is to train the well generalized, unbiased model in TensorFlow.The objectives for the proposed solution are enlisted below:Real-Time predictions with non-existent latency.Generating alerts for the driver in real time, so the overall safety of the system can be enhanced.Training the model in well generalized fashion, capable of classifying with high degree of accuracy in challenging environment.
