loader animation

Driving Fatigue Classification: Begal And Eeg Signals

 Feb 3, 2023
 BEGAL and EEG Signals

Driving Fatigue Classification: BEGAL and EEG Signals

Modern society's fast-paced lifestyle not only encourages society's quick development but also puts more stress on employees' ability to complete their jobs. The mental tiredness of workers can readily be brought on by prolonged high-intensity employment. Besides lowering productivity at work, mental exhaustion is bad for your health and can even increase the risk of accidents. One of the major factors contributing to road accidents in recent years is fatigued driving. Long-haul and overtime truck drivers are frequently prone to exhaustion or drowsiness, and some fall asleep behind the wheel. As a result, the problem of fatigue detection has gained importance in driving safety. Scientists use this information to develop pertinent rules and attempt to lessen the incidence and severity of road accidents.


The development of technologies for the identification of driving weariness is currently plagued by numerous issues. For instance, the method of measuring the pupil diameter is challenging to execute, the process of detecting the physical reaction of the driver has a high false alarm rate, and the method of detecting vehicle parameters needs clear markings on the road's center line and good lighting. Diagnosing fatigue via biological signals is very alluring because of the significant investments made in recent years by scientific researchers in the study of non-contact ECG and EEG collecting. It is a foundation for further research and creating onboard, real-time driving tiredness alert devices. It also gives traffic management agencies a solid foundation upon which to act to combat driving fatigue and reduce traffic accidents caused by human error. (Dr. Mu)


Currently, the diagnosis of mental fatigue, both domestically and internationally, focuses mostly on the following three traits:


  1. Behavioral characteristics: Behavioral traits one of the often utilized driving indications to detect weariness through vehicle behavior, including variations in lateral lane locations and variability in vehicle heading disparities. Although this strategy is simple to comprehend, it is difficult to produce steady and reliable recognition results because the standards for these qualities have not yet been established.

  2. Facial expression characteristics: Facial expression traits include things like how often you nod and how often your eyes close. Although this method is incredibly user-friendly, there are some variables, such as image angle and brightness, that might alter the recognition of these features. These drawbacks diminish this method's overall recognition accuracy and restrict its use cases.

  3. Physiological characteristics: Among the physiological traits are the electrocardiogram (ECG), heart rate, electromyogram (EMG), and EEG. These features, which map the physiological changes of the human body, differ from behavioral and facial features in that they accurately reflect the body's present condition and environment. As a result, physiological feature-based driving fatigue detection is a very popular application area. The most straightforward, efficient, and promising way to identify driving weariness among these physiological feature-based methods is the monitoring of EEG waves. (Dr. Wang)

Fatigue is a complex physiological phenomenon caused by prolonged or intense physical or mental work that throws the body's physiological and psychological systems out of equilibrium. Driving fatigue is caused by the driver's excessive consumption of lengthy periods of time spent driving or engaging in other strenuous physical activity or by lack of sleep, which results in drowsiness, sluggishness, and limb weakness. Phenomena in the context. Numerous characteristics, both visible and invisible, such as driving posture, blink rate, eye closure time, nodding action, reaction time, facial expressions, and EEG, ECG, and EMG signal, as well as skin temperature, skin resistance, breathing rate, and others, will vary in a tired driver. The main symptoms include drowsiness and weakness of the driver, difficulty overcoming drowsiness, decreased vision, a gradually narrowing field of vision, inattention, decreased judgment, decreased thinking ability, weak limbs, decreased sensory organ function, erratic driving movements, delayed assessment, rhythm disorders, fatigue in the knees, stiffness in the neck, backache, and legs, and loss of self-control, leading to several uncomfortable symptoms like being unable to concentrate.

Shifting could be more precise and timely when the driver is tired. When operating under considerable exhaustion, the process is slow and occasionally even forgets to run. In extreme fatigue, driving loss of control is frequently caused by subconscious behavior or a short-term sleep phenomenon.

Spotting a driver's condition of exhaustion has become a prominent topic in recent years. An EEG-based driver fatigue test procedure is defined by Chain R. The findings suggest that EEG frequency features can be effectively used for driving tiredness testing, but additional research is required before the study's overall conclusions can be drawn. In driving simulation tests, Yimyam W employed 50 recorded EEGs to produce the two exhaustion states of alertness and drowsiness. According to the experimental findings, the two complexity parameters considerably decrease as the degree of fatigue rises. According to the results, these two nonlinear indicators can be used to define driver weariness. However, his study's lack of experimental data caused minor variations in the sample sets, producing false results. Zeng researched how to evaluate EEG signals' complicated, erratic, and nonlinear properties. The findings demonstrate that this research uses a mix of finite element features and the AdaBoost classifier to identify driver weariness based on EEG, providing the assurance needed to investigate internal physiological mechanisms and wearable applications. The experiment did not account for the subject's gender and age deviations from the established parameters. (Dr. Zeng)