Implementation of driver's drowsiness assistance model based on eye movements detection

Jumana Waleed,Thekra Abbas,Taha Mohammed Hasan

doi:10.15587/1729-4061.2020.211755

Abstract

Annually, lots of persons are carrying lifelong disabilities or losing their lives owing to fatal accidents on the road. In addition to mechanical failures and people errors, driver's drowsiness represents one of the fundamental reasons for fatal accidents on the road. When drivers feel drowsy, lots of physiological and behavioral symptoms appear such as changes in the waves of the human brain, changes in facial expressions, variations in eye activity, decreasing head movements, etc. Therefore, there is a significant necessity to provide developed models of driver's drowsiness detection that exploit these symptoms for reducing accidents by warning drivers of drowsiness and fatigue. This paper concentrates on proposing a driver's drowsiness assistance model to monitor and alarm drivers by utilizing a behavioral-based method (eye movements detection method). In the proposed method of detecting eye movements (closed/opened), the Advanced Local Binary Pattern (Advanced LBP) is used, in which the descriptors are utilized to represent eye images to extract the tissue features of different persons in the driving car to see if the driver is in a drowsy state or not and this occurs after recording the driver's video and detecting the eyes of the driver. To extract the features in this way, the image of the eyes is divided into small regions through the Advanced LBP and sequenced into a single feature vector, where this method is used to determine the similarity features in the training group and to classify the eye image. The Naive Bayes classifier (NB) and Support Vector Machine (SVM) are utilized for giving good accuracy. The results indicate that the system has a high accuracy rate compared with the other existing methods, where the accuracy rate of NB and SVM using an eye detection dataset with training 90 % and testing 10% is 96 % and 97 %, respectively

Highlights

Driving is a complicated job, which requires mental awareness as well as physical and behavioral resources
The need for complete mental awareness makes it a risky job since people have limited ability to become mindful for a long time
Classification of eye state detection The proposed model utilized two of the most popular machine learning classifiers (NB and Support Vector Machines (SVM)) to decide if the person is in a drowsy state or not

Summary

Introduction

Driving is a complicated job, which requires mental awareness as well as physical and behavioral resources. Physiological-based methods provide a precise and objective manner for sleepiness measuring by utilizing different physiological signals of the human body, like electrooculogram (EOG), electroencephalogram (EEG), and electrocardiogram (ECG) [4] These methods depend on the concept that physiological signals begin to vary in the earlier drowsiness stages, which gives the system of a possible driver’s drowsiness detection a little additional time. The behavioral-based methods utilize a video camera for acquiring images based on an integration of machine learning and computer vision approaches for detecting events of interest, measuring them, making a decision if the driver is drowsy or not. Studies concentrate on the process of detecting the eyes state to specify whether a driver is drowsy Most of these studies worked on extracting features, followed by training and utilization of machine learning algorithms of distinct abilities, weaknesses, and strengths

Literature review and problem statement

Experimental results

Discussion of the proposed system

Selecting effective classifiers led to the high accuracy