Published Paper Details:

In order to drive the state of the art technology for Driver's safety, the demand for Artificial Intelligence and Machine learning delicacy approaches are popularly used in ultramodern Classrooms. This paper presents a Driver drowsiness detection architecture that may be placed within the vehicle as a standalone design unit. In this organized deep learning CNN framework, a comprehensive analytics strategy is employed to detect somnolence in automobilists. There are two distinct phases employed in this experimental framework namely image data acquisition followed by identification of the region of interest (ROI) in phase 1. In the coming phase, the deep learning algorithm is employed to classify and prognosticate the result. The object detection training is using the Haar Cascade Classifier which facilitates getting an accurate model. The model is trained by exploiting the open-source tools namely Keras alongside Tensorflow as backend. The hardware design includes Raspberry pi model 4 or 3b, camera, GSM Module, and a speaker. The video capturing, labeling of open or closed eyes and the score are attained as a real-time status with the speaker output to warn the automobilist during sleepiness state. The standalone hardware module is uploaded with the entire training AI software. Real-time image acquisition by the design platform and forecasting the driver drowsiness detection, depicts a delicacy of 95 to 96. The activation function ReLU used during the predicting state in CNN improves the computational effectiveness of the design My SQL model.