Driver Drowsiness Detection using Microservices and Convolutional Neural Network

Road accidents are one of the main contributors to net fatality rates in India. According to a recent survey in 2020, 43% of road accidents come from drowsy driving. Driving over hours and being in the same state makes the driver feel exhausted and fatigue leading them to drowsiness. A report from Road Transport of India stated that on average 5210 tragedies occur each year alone on the highways of India. A primary system to measure and alert the driver must be mandatory for any moving vehicle. In this paper, a modern approach is proposed for real-time drowsiness detection. A production-grade application with microservice architecture is one of the main focus of this paper. The process of building up the data, augmenting it to a desired level and finally labeling is presented. The customized state of art model is proposed that can achieve an accuracy of 83.65%. Keywords—Deeplearning; microservices; drowsiness detection system; real-time application; kubernetes