CNN Model to Build Drowsiness Detection System

CNN Model to Build Drowsiness Detection System

Prof. D. R. Uike,

Anurag Shrirao, Kunal Raut, Pratik Moharkar, Yash Lanjewar

(Department of Information Technology, Government College of Engineering, Amravati – 444604, Maharashtra, India)

(dilip.uike100, anuragshrirao, kunalraut2110, pratikmoharkar, yashlajewar01)@gmail.com

ABSTRACT: Travel during rush hour or on holiday, uncover the driver to a traffic jam for an hour, thus causing the driver to feel drowsy easily due to high concentration and lack of rest. In this situation, it increases the percentage of a car accidents due to car driver overtiredness is the primary origin of the car accident. So, this drowsiness detection system is proposed to detect the state of the car driver using techniques like face detection, drowsiness is defined to be a condition of exhaustion, where the expression of the face is different from usual. The important steps in detecting drowsiness are face detection, eye detection, and expression detection. The light and position of the camera is the major problems. In this project, different architectures such as haarcascade classifier and CNN were used to analyze the performance of face and drowsiness detection. Also, we have proposed new detection methods using deep learning techniques. To detect the driver’s state, we use facial regions corresponding to the entire face. For the Classification, The CNN (Convolutional Neural Network) architecture integrated in the drowsiness detection.

Keywords: Convolutional Neural Network (CNN), Drowsiness detection, Face Detection, Deep Learning, multi-modal analysis