Driver Drowsiness Detection and Monitoring System using Machine Learning
Authors:
Open Access
- 1 August 2022
- conference paper
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 2325 (1)
Abstract
In the present world, lots of road accidents take place due to the lack of attention and alertness of driver. This is termed as driver drowsiness. This leads to a lot of unfortunate situations causing adverse damage to human lives. The main goal of this research is the detection of driver drowsiness and an appropriate response to the detection. There are many methods which are based on the motion of the vehicle or based on the driver’s behavior. One of the methods is the physiological method which helps in distracting the driver from drowsiness and making him alert. And few methods require expensive sensors and deals with a lot of data. Therefore, this paper develops a system for detecting drowsiness in real time with proper procedure and accuracy which is acceptable. In this system, the driver’s facial expressions are captured and recorded using a webcam. Every movement in each frame is detected using few techniques of image processing. The Eye Aspect Ratio, Mouth Opening Ratio, and Nose Length Ratio are calculated using the landmark points on the face. The calculated values are compared to the threshold values developed by the system and the difference in value leads to the detection. At the same time, the machine learning algorithms are also implemented in offline manner. Based on the classification, the system has successfully achieved 95.58% of sensitivity and 100% of specificity using Support Vector Machine. This model system is compatible with all kinds of vehicles.Keywords
This publication has 11 references indexed in Scilit:
- Driver Drowsiness Detection2020
- FPGA-based object detection processor with HOG feature and SVM classifier2019
- Driver Face Recognition and Sober Drunk Classification using Thermal Images2019
- Real Time Driver Fatigue Detection Based on SVM Algorithm2018
- Decision Tree Ensemble Vs. N.N. Deep Learning: Efficiency Comparison For A Small Image Dataset2018
- A Multimodal System for Assessing Alertness Levels Due to Cognitive LoadingIEEE Transactions on Neural Systems and Rehabilitation Engineering, 2017
- An Accurate ECG-Based Transportation Safety Drowsiness Detection SchemeIEEE Transactions on Industrial Informatics, 2016
- Comparative study of Principal Component Analysis based Intrusion Detection approach using machine learning algorithms2015
- IEEE Smart Grid Vision for Vehicular Technology: 2030 and Beyond2014
- Estimation and identification of two-dimensional imagesIEEE Transactions on Automatic Control, 1983