Driver Fatigue Research Articles

IoT-Assisted Automatic Driver Drowsiness Detection through Facial Movement Analysis Using Deep Learning and a U-Net-Based Architecture

The main purpose of a detection system is to ascertain the state of an individual’s eyes, whether they are open and alert or closed, and then alert them to their level of fatigue. As a result of this, they will refrain from approaching an accident site. In addition, it would be advantageous for people to be promptly alerted in real time before the occurrence of any calamitous events affecting multiple people. The implementation of Internet-of-Things (IoT) technology in driver action recognition has become imperative due to the ongoing advancements in Artificial Intelligence (AI) and deep learning (DL) within Advanced Driver Assistance Systems (ADAS), which are significantly transforming the driving encounter. This work presents a deep learning model that utilizes a CNN–Long Short-Term Memory network to detect driver sleepiness. We employ different algorithms on datasets such as EM-CNN, VGG-16, GoogLeNet, AlexNet, ResNet50, and CNN-LSTM. The aforementioned algorithms are used for classification, and it is evident that the CNN-LSTM algorithm exhibits superior accuracy compared to alternative deep learning algorithms. The model is provided with video clips of a certain period, and it distinguishes the clip by analyzing the sequence of motions exhibited by the driver in the video. The key objective of this work is to promote road safety by notifying drivers when they exhibit signs of drowsiness, minimizing the probability of accidents caused by fatigue-related disorders. It would help in developing an ADAS that is capable of detecting and addressing driver tiredness proactively. This work intends to limit the potential dangers associated with drowsy driving, hence promoting enhanced road safety and a decrease in accidents caused by fatigue-related variables. This work aims to achieve high efficacy while maintaining a non-intrusive nature. This work endeavors to offer a non-intrusive solution that may be seamlessly integrated into current automobiles, hence enhancing accessibility to a broader spectrum of drivers through the utilization of facial movement analysis employing CNN-LSTM and a U-Net-based architecture.

MIDGNWD : Design of an efficient multimodal iterative deep graph network wearable interface for detecting drowsy drivers

Drowsy driving is a significant contributor to road traffic accidents, necessitating the development of robust and efficient detection systems. Existing approaches for detecting drowsy drivers often rely on single modalities such as image-based facial analysis or vehiclebased metrics, which may lack precision and be susceptible to environmental variations. Furthermore, the existing models might be computationally expensive and result in delayed responses. To address these limitations, this paper proposes a novel Multimodal Iterative Deep Graph Network Wearable Interface that leverages a rich set of data including heart rate, respiration, accelerometer movement, location (speed), traffic, and weather data. These data samples are processed through a fusion of different Recurrent Neural Network processes. The Bidirectional Long Short-Term Memory (BiLSTM) which is used to generate augmented features, is fused with Bidirectional Gated Recurrent Unit (BiGRU) which assists in estimating multidomain feature sets. These features are then used to train an efficient Deep Graph Recurrent Network (DGRN), which facilitates the real-time identification of drowsy drivers. The proposed model significantly enhances the precision of drowsy driver detection by 8.5%, accuracy by 9.5%, recall by 8.3%, and Area Under the Curve (AUC) by 4.9%, while also improving specificity by 5.5% and reducing delay by 10.4%. These improvements collectively contribute to a substantial enhancement in road safety, potentially saving lives by preventing accidents due to drowsy driving. The proposed system, with its real-time processing capability and high accuracy, stands as a significant improvement for identifying drowsy drivers, paving the way for safer and smarter transportation systems.

EEG Information Transfer Changes in Different Daily Fatigue Levels During Drowsy Driving.

A significant issue for traffic safety has been drowsy driving for decades. A number of studies have investigated the effects of acute fatigue on spectral power; and recent research has revealed that drowsy driving is associated with a variety of brain connections in a specific cortico-cortical pathway. In spite of this, it is still unclear how different brain regions are connected in drowsy driving at different levels of daily fatigue. This study identified the brain connectivity-behavior relationship among three different daily fatigue levels (low-, median- and high-fatigue) with the EEG data transfer entropy. According to the results, only low- and medium-fatigue groups demonstrated an inverted U-shaped change in connectivity from high performance to poor behavioral performance. In addition, from low- to high-fatigue groups, connectivity magnitude decreased in the frontal region and increased in the occipital region. These study results suggest that brain connectivity and driving behavior would be affected by different levels of daily fatigue.

ТЕНДЕНЦИИ ИЗМЕНЕНИЯ ФУНКЦИОНАЛЬНОГО СОСТОЯНИЯ ВОДИТЕЛЯ-ХОЛЕРИКА В ДОРОЖНОМ ЗАТОРЕ

In this article, as an object of research, the process of a driver's work activity is considered when performing his functions in the urban transport system. Studies have been conducted to assess the impact of traffic congestion on the functional state of a choleric driver, which is one of the widespread and well-known types of temperament. The hypothesis of the study is that a change in the functional state of a driver in a traffic jam affects the degree of fatigue and depends on the condition of the driver and the parameters of the congestion. To assess the level of fatigue of the driver, the concept of the cardiovascular system is used as an appropriate mechanism of adaptive activity of the body. At the same time, an electrocardiogram is measured and the irregularities of cardiac intervals are analyzed, which are a universal response to all types of loads. The degree of fatigue level of the driver is measured in conventional units and is determined by a special technique taking into account statistical indicators, histogram indicators and spectral analysis of cardiac intervals. Using a nonlinear regression model of changes in the functional state of a choleric driver, the corresponding patterns of changes in the level of fatigue for traffic jams of different duration have been determined. As a result of the analysis of the constructed graphs of changes in the final level of fatigue of choleric drivers of different age categories in congestion, it was found that the most significant factor is the value of the initial level of fatigue of the driver. The second most important factor influencing the change in the final level of fatigue of a choleric driver is the duration of traffic congestion. It was revealed that this factor affects the degree of fatigue of the driver at the end of the jam only together with the initial level of fatigue. The age of the choleric driver was in third place in terms of the importance of influencing his final level of fatigue in congestion. It was also found that older drivers, compared with younger drivers, experience the negative impact of traffic congestion with an increase in its duration. Studies have shown that traffic jams lasting more than twelve minutes lead to a significant increase in the final level of fatigue of the choleric driver. This may increase the likelihood of a traffic accident. The established trends in the functional state of the choleric driver in traffic congestion allow us to predict his behavior after congestion and evaluate possible patterns of development of the road traffic situation.

A sophisticated Drowsiness Detection System via Deep Transfer Learning for real time scenarios

<abstract> <p>Driver drowsiness is one of the leading causes of road accidents resulting in serious physical injuries, fatalities, and substantial economic losses. A sophisticated Driver Drowsiness Detection (DDD) system can alert the driver in case of abnormal behavior and avoid catastrophes. Several studies have already addressed driver drowsiness through behavioral measures and facial features. In this paper, we propose a hybrid real-time DDD system based on the Eyes Closure Ratio and Mouth Opening Ratio using simple camera and deep learning techniques. This system seeks to model the driver's behavior in order to alert him/her in case of drowsiness states to avoid potential accidents. The main contribution of the proposed approach is to build a reliable system able to avoid false detected drowsiness situations and to alert only the real ones. To this end, our research procedure is divided into two processes. The offline process performs a classification module using pretrained Convolutional Neural Networks (CNNs) to detect the drowsiness of the driver. In the online process, we calculate the percentage of the eyes' closure and yawning frequency of the driver online from real-time video using the Chebyshev distance instead of the classic Euclidean distance. The accurate drowsiness state of the driver is evaluated with the aid of the pretrained CNNs based on an ensemble learning paradigm. In order to improve models' performances, we applied data augmentation techniques for the generated dataset. The accuracies achieved are 97 % for the VGG16 model, 96% for VGG19 model and 98% for ResNet50 model. This system can assess the driver's dynamics with a precision rate of 98%.</p> </abstract>

A Novel Brain-Machine Safety System Based on Drowsiness Detection using the PNN and MLP Algorithms

This work aims to develop a safety system capable of detecting drowsiness, in order to reduce the accident rate, the main idea is based on the classification of the cerebral signal captured by the electroencephalograph, after processing to eliminate all noise, the resulting data will then be segmented into two classes, the first category is devoted to the state of wakefulness and the second will be devoted to drowsiness. This categorization is based on the combination of features such as mean and standard deviation, and performed by two neural network algorithms, the probabilistic neural network and the multilayer perceptron, whose performance is evaluated according to machine learning metrics such as accuracy, error, specificity, sensitivity, F1 indicator and ROC curve representation. The binary classification results are highly significant in terms of class discrimination, with both models achieving excellent performance, with the MLP classifier outperforming the PNN model with an accuracy of 96.10% and an accuracy rate of 98.43%. This suggests that the MLP model can be effectively applied to critical tasks where good discrimination is required, such as detecting drowsiness in drivers to avoid accidents.

К ВОПРОСУ ОБ ИЗМЕНЕНИИ ВРЕМЕНИ РЕАКЦИИ ВОДИТЕЛЯ В ТРАНСПОРТНОМ ПОТОКЕ ВЫСОКОЙ ПЛОТНОСТИ

This article considers the problem of determining the reaction time of sanguine drivers of different age categories in the city's transport system in conditions of high-density traffic flow, close to congestion. Prolonged exposure to high-density traffic leads to a deterioration in the functional state of the driver and to an increase in his psycho-emotional tension. A change in the functional state causes fatigue in the driver, which in turn leads to an increase in his reaction time, which affects road safety. The assessment of the driver's fatigue level was carried out by measuring an electrocardiogram and analyzing the unevenness of cardiac intervals, which are a universal response to all types of loads. A model for estimating changes in the reaction time of a sanguine driver based on a regression model of changes in his fatigue level is proposed. The graphs show the changes in the reaction time of drivers of different ages during their stay in traffic congestion. It was revealed that the reaction time of the driver in the first minutes of being in a traffic jam changes slightly, due to the presence of reserve compensatory capabilities of the body, however, after the fifth to sixth minute, an objective process of deterioration of the functional state begins. It was found that the change in the reaction time of a sanguine driver in a traffic jam primarily depends on the initial level of fatigue. When the driver stays in a high-density traffic flow with the initial state of optimal voltage of the regulatory systems, the change in his reaction time by the end of the jam reaches the greatest value. It was also revealed that the greatest changes in the reaction time of a sanguine driver occur in older drivers.

Driver Drowsiness Detection using Evolutionary Machine Learning: A Survey

One of the factors that kills hundreds of people every year is driving accidents caused by drowsy drivers. There are different methods to prevent this type of accidents. Recently Machine Learning (ML) and Deep Learning (DL) have emerged as very effective and valuable approaches for detecting driver drowsiness. Moreover, the optimization of machine learning (ML) and deep learning (DL) models may be achieved through the utilization of evolutionary algorithms (EA). This survey aims to offer an overview of recent studies in driver drowsiness detection-based machine learning and deep learning models that have been improved by EA. This survey divides the approaches for detecting drowsiness into two groups: those that rely on ML, and DL, and those that rely on models-based deep learning and machine learning that are optimized by evolutionary algorithms.

Effect of Split Condition and Driving Duration on Fatigue Based on Changes in Brain Wave Signals when Driving a Train Simulator

One of the causes of train accidents is driver fatigue. Train driver fatigue can be caused by sleep factors, known as split sleep. This study aims to assess the impact of split sleep on train driver fatigue. A total of 12 male participants were asked to drive a train simulator for 2.5 hours after facing two sleep conditions, namely split sleep and baseline. The split sleep condition required participants to sleep in two segments at 05.00-10.00 and 12.00-15.00, while the baseline condition was conducted in one segment at 21.00-05.00. Fatigue was measured based on changes in brain wave signals via electroencephalogram (EEG) and Swedish Occupational Fatigue Inventory (SOFI). Fatigue measurements with EEG were conducted at the 10-minute start and end of the simulation, while fatigue measurements with SOFI were conducted before and after the simulation. The results of this study showed a higher level of subjective fatigue in split sleep compared to the baseline. However, the EEG signal change data and other dimensions of SOFI dimensions showed no difference between the two sleep states. Another result was an increase in fatigue after simulation in all dimensions of the SOFI. Therefore, split sleep should not be applied by drivers because it can increase subjective fatigue. However, if split sleep needs to be applied, it is necessary to fulfill sleep quantity (7-9 hours) and improve sleep quality. In addition, the company also needs to ensure that the train driver are awake at least 15 minutes.

Prediction of Dangerous Driving Behaviour Based on Vehicle Motion

The integration of deep learning, computer vision, and advanced algorithms has ushered in a transformative era in the prediction of human driving behavior, consequently revolutionizing road safety. This paper focuses on an innovative convergence of technology that addresses critical issues like driver fatigue and distracted driving by automatically identifying and categorizing diverse driving behaviors. Neural network architectures, such as VGG16, AlexNet, and ResNet are described in this paper that have propelled accuracy in behavior classification to remarkable levels. However, the quest for safer roads remains ongoing, with promising avenues lying ahead. First and foremost, the creation of extensive, diverse, and meticulously annotated datasets is paramount. These datasets serve as the bedrock upon which future models can be trained, enhancing their robustness and generalizability across a spectrum of driving scenarios. Real-time models represent another pivotal frontier. These models hold the potential to provide timely interventions and support systems for drivers, thus preventing accidents proactively. The exploration of hybrid techniques that amalgamate the strengths of various neural network architectures presents an exciting avenue, promising to further push the boundaries of prediction accuracy. Furthermore, this paper also discusses the fusion of multi-modal data, encompassing sensor data from IoT and smartphone devices, that holds immense promise. This holistic approach promises a more comprehensive understanding of driver behavior by integrating diverse data sources, ultimately contributing to the creation of safer road environments.In this research paper, we also explore these cutting-edge developments in deep learning and computer vision, emphasizing technical novelty and innovation. Through an interdisciplinary approach, we envision a future where the synergy of technology, data, and human behavior leads to a substantial reduction in road accidents and improved road safety.

Fast eye centre localization using combined unsupervised technics

Eye movements offer precious information about persons? state. Video surveillance, marketing, driver fatigue as well as medical diagnosis assistance applications manage eye behavior. We propose a new method for efficiently detecting eye movement. In this paper, we combine circle eye model with eye feature method to improve the accuracy. A set of detectors estimate the eyes centers to increase the localization rate. As a pre-processing stage, the mean of the edges yields the center of the two eye regions. Image treatment operations reduce the ROI. A Circle Hough Transform (CHT) algorithm is adopted in a modified version as a detector to find the circle eye in the image; the circle center found represents the eye's pupil estimation. We introduced the Maximally Stable Extremal Region (MSER) as a second detector, which has never been used for eye localization. Invariant to continuous geometric transformations and affine intensity changes and detected at several scales, MSERs efficiently detect regions of interest, in our case eye regions, and precisely, their centers. Ellipses fit MSERs, and their centroid estimation match eyes center. We demonstrate that the true eye centers can be found by combining these methods. The validation of the proposed method is performed on a very challenging BioID base. The proposed approach compares well with existing state-of-the-art techniques and achieves an accuracy of 82.53% on the BioID database when the normalized error is less than 0.05, without prior knowledge or any learning model.

An Overview of Machine Learning Algorithms to Reduce Driver Fatigue and Distraction-Related traffic Accidents

Sensor-equipped driver monitoring systems are a significant advancement in vehicle safety technology, continuously monitoring drivers’ condition and behavior using various sensors. These systems identify exhaustion, inattention, or impairment indicators to improve road safety. They can measure the driver’s focus and attentiveness by tracking their head position, eye movements, and facial expressions. If detected, the system can notify the driver via visual, aural, or tactile alerts, prompting the driver to refocus or take a break. Some sophisticated systems can even take over the car’s steering in emergencies. The study focuses on integrating artificial intelligence (AI) and machine learning (ML) into Driver Monitoring Systems (DMS) to reduce accidents caused by driver exhaustion and distraction. The study discusses two primary DMS systems: Multi-Sensor Based Systems (S-HDx) and Vision-Based Systems (V-HDx). Vision-based systems examine drivers or specific driving characteristics to identify whether a driver is inattentive or sleepy. Multi-sensor-based systems identify drowsiness using physiological or non-visual factors. The study evaluated AI and ML algorithms for fatigue and distraction detection in drivers. Algorithms used in vision-based systems include single-shot detection, MobileNetV2, feature pyramid networks, and convolutional neural networks. Multi-sensor-based systems use algorithms such as SVM, CNN, XGBoost, and decision trees. Vision-based systems are recommended for DMS development due to their user-friendliness and non-intrusive nature. Future research could independently examine these techniques and algorithms to create a more successful DMS.

Study on a Portable Electrode Used to Detect the Fatigue of Tower Crane Drivers in Real Construction Environment

Study on a Portable Electrode Used to Detect the Fatigue of Tower Crane Drivers in Real Construction Environment

The architecture of functional brain network modulated by driving under train running noise exposure.

A noisy environment can considerably impact drivers' attention and fatigue, endangering driving safety. Consequently, this study designed a simulated driving experimental scenario to analyse the effects of noise generated during urban rail transit train operation on drivers' functional brain networks. The experiment recruited 16 participants, and the simulated driving scenario was conducted at noise levels of 50, 60, 70, and 80 dB. Functional connectivity between all electrode pairs across various frequency bands was evaluated using the weighted phase lag index (WPLI), and a brain network based on this was constructed. Graph theoretic analysis employed network global efficiency, degree, and clustering coefficient as metrics. Significant increases in the WPLI values of theta and alpha frequency bands were observed in high noise environments (70 dB, 80 dB), as well as enhanced brain synchronisation. Furthermore, concerning the topological metrics of brain networks, it was observed that the global efficiency of brain networks in theta and alpha frequency ranges, as well as the node degree and clustering coefficients, experienced substantial growth in high noise environments (70 dB, 80 dB) as opposed to 50 dB and 60 dB. This finding indicates that high-noise environments impact the reorganisation of functional brain networks, leading to a preference for network structures with improved global efficiency. Such findings may improve our understanding of the neural mechanisms of driving under noise exposure, and thus potentially reduce road accidents to some extent.

Cognitive Aspect of Emotion Estimation of a Driver

Despite rapid advancements in the automotive industry, traffic safety risks persist. Addressing this challenge requires innovative driver assistance technologies. Common accidents result from driver inattention, fatigue, and stress, leading to issues like falling asleep at the wheel and improper acceleration and braking. Our study aims to contribute to advanced driver assistance systems that adapt to drivers' emotional needs, ultimately enhancing road safety. In this paper, we mention the result of our research to estimate drivers' emotions using sensors. For that purpose, we developed a sensor network containing sensors such as EEG, eye tracker, and driving simulator. We explored the relationship. As a result, we confirmed the relation between the driver's emotions, especially sleep conditions, driving speed, duration, and brain wave behavior.

DriveAlert: Improving Road Safety with Real-time Monitoring of Driver Drowsiness through Blink Rate

DriveAlert: Improving Road Safety with Real-time Monitoring of Driver Drowsiness through Blink Rate

A Connectivity-Aware Graph Neural Network for Real-Time Drowsiness Classification.

Drowsy driving is one of the primary causes of driving fatalities. Electroencephalography (EEG), a method for detecting drowsiness directly from brain activity, has been widely used for detecting driver drowsiness in real-time. Recent studies have revealed the great potential of using brain connectivity graphs constructed based on EEG data for drowsy state predictions. However, traditional brain connectivity networks are irrelevant to the downstream prediction tasks. This article proposes a connectivity-aware graph neural network (CAGNN) using a self-attention mechanism that can generate task-relevant connectivity networks via end-to-end training. Our method achieved an accuracy of 72.6% and outperformed other convolutional neural networks (CNNs) and graph generation methods based on a drowsy driving dataset. In addition, we introduced a squeeze-and-excitation (SE) block to capture important features and demonstrated that the SE attention score can reveal the most important feature band. We compared our generated connectivity graphs in the drowsy and alert states and found drowsiness connectivity patterns, including significantly reduced occipital connectivity and interregional connectivity. Additionally, we performed a post hoc interpretability analysis and found that our method could identify drowsiness features such as alpha spindles. Our code is available online at https://github.com/ALEX95GOGO/CAGNN.

Driver Drowsiness Detection Based on Joint Human Face and Facial Landmark Localization With Cheap Operations

Driver Drowsiness Detection Based on Joint Human Face and Facial Landmark Localization With Cheap Operations

A Systematic Review on Driver Drowsiness Detection Using Eye Activity Measures

A Systematic Review on Driver Drowsiness Detection Using Eye Activity Measures

Developing A method to Detect Driver Drowsiness Based on A single EEG Channel and Discriminated Features

Developing A method to Detect Driver Drowsiness Based on A single EEG Channel and Discriminated Features