Driver drowsiness detection based on non-intrusive metrics considering individual specifics

Wang, Xuesong; Xu, Chuan

doi:10.1016/j.aap.2015.09.002

Cited by 138 publications

(102 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Physiological methods, including EEG, ECG, and electrooculogram (EOG), are regarded as the most accurate DDD approaches [14][15]17,[28][29]. A major contribution to their high accuracy is their capability of gaining direct access to drivers' inner body electrical signals, which can, however, be used for immediate detection of alertness changes [30].…”

Section: Problem Backgroundmentioning

confidence: 99%

“…EEG and ECG are used for tracking the driver's brain activity [8][9][10] and heart pulse rate [11][12][13] respectively. Although both approaches offer higher accuracy, they are actually not feasible in the real-world [14]. This is because they require electrodes being attached to the driver, which could cause discomfort as well as distractions [15,16].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Non-intrusive Driver Drowsiness Detection based on Face and Eye Tracking

Bamidele¹,

Kamardin²,

Aziz³

et al. 2019

IJACSA

View full text Add to dashboard Cite

The rate of annual road accidents attributed to drowsy driving are significantly high. Due to this, researchers have proposed several methods aimed at detecting drivers' drowsiness. These methods include subjective, physiological, behavioral, vehicle-based, and hybrid methods. However, recent reports on road safety are still indicating drowsy driving as a major cause of road accidents. This is plausible because the current driver drowsiness detection (DDD) solutions are either intrusive or expensive, thus hindering their ubiquitous nature. This research serves to bridge this gap by providing a test-bed for achieving a non-intrusive and low-cost DDD solution. A behavioral DDD solution is proposed based on tracking the face and eye state of the driver. The aim is to make this research an inception to DDD pervasiveness. To achieve this, National Tsing Hua University (NTHU) Computer Vision Lab's driver drowsiness detection video dataset was utilized. Several video and image processing operations were performed on the videos so as to detect the drivers' eye state. From the eye states, three important drowsiness features were extracted: percentage of eyelid closure (PERCLOS), blink frequency (BF), and Maximum Closure Duration (MCD) of the eyes. These features were then fed as inputs into several machine learning models for drowsiness classification. Models from the K-nearest Neighbors (KNN), Support Vector Machine (SVM), Logistic Regression, and Artificial Neural Networks (ANN) machine learning algorithms were experimented. These models were evaluated by calculating their accuracy, sensitivity, specificity, miss rate, and false alarm rate values. Although these five metrics were evaluated, the focus was more on getting optimal accuracies and miss rates. The result shows that the best models were a KNN model when k = 31 and an ANN model that used an Adadelta optimizer with 3 hidden layer network of 3, 27, and 9 neurons respective. The KNN model obtained an accuracy of 72.25% with a miss rate of 16.67%, while the ANN model obtained 71.61% and 14.44% accuracy and miss rate respectively.

show abstract

Section: Problem Backgroundmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Non-intrusive Driver Drowsiness Detection based on Face and Eye Tracking

Bamidele¹,

Kamardin²,

Aziz³

et al. 2019

IJACSA

View full text Add to dashboard Cite

show abstract

“…There is the ongoing research on mobile devices that can be built into cars and analyze some physiological parameters (e.g., percentage of eyelid closure, blink velocitybased score, average pupil diameter, electroencephalogram (EEG) signal, standard deviation of lateral position and steering wheel reversals) to warn a driver in advance [24][25][26].…”

Section: Discussionmentioning

confidence: 99%

Reliability of the Epworth Sleepiness Scale and the Berlin Questionnaire for screening obstructive sleep apnea syndrome in the context of the examination of candidates for drivers

Kiciński

Przybylska-Kuć

Tatara³

et al. 2016

Med Pr

View full text Add to dashboard Cite

Background: The aim of the study has been to assess the usefulness of the Epworth Sleepiness Scale (ESS) and the Berlin Questionnaire (BQ) for obstructive sleep apnea syndrome (OSAS) screening. The capacity of both tests to discriminate between healthy individuals or with mild OSAS (apnea-hypopnea index (AHI) < 15/h) vs. patients with moderate or severe OSAS (AHI ≥ 15/h) was evaluated. Material and Methods: The study encompassed 223 patients with a suspicion of the OSAS. The ESS and BQ were completed by patients unassisted. Screening polysomnography was performed using the Porti SleepDoc. The OSAS was diagnosed when AHI ≥ 15/h or AHI ≥ 5/h with simultaneous occurrence of clinical symptoms. Results: The ESS score was found to be significantly higher in the study group compared to the control group (8.9±5.9 vs. 11.6±5.2 pt, p < 0.0001). Otherwise, there were no significant inter-group differences in the percentage of high-risk individuals according to the BQ (83.7% vs. 92.3%, p > 0.05). Sensitivity of the ESS and BQ was 53.2% and 93.1%, respectively while specificity was 58.8% and 16.2%, respectively. Poor correlation between the ESS score and AHI and apnea index were noticed (r = 0.22, p = 0.001 and r = 0.24, p < 0.001, respectively). Conclusions: Considering its low sensitivity, the ESS should not be used as a screening test for the OSAS diagnosis amongst candidates for drivers. The BQ is characterised by high sensitivity for the OSAS diagnosis with AHI ≥ 15/h, however, due to low specificity, the questionnaire may increase the number of healthy individuals referred for needless diagnostic procedures. Med Pr 2016;67(6):721-728

show abstract

“…In China, 1768 fatalities were attributed to fatigue driving in 2007. Previous research has reported that fatigue was responsible for 20-30% of the total road fatalities [1,2]. In the following section, we present the state of the art of enhancement techniques using computerized machine.…”

Section: Introductionmentioning

confidence: 99%

Vigilance Enhancement Using Computerized Techniques

Al-Shargie¹

2019

Preprint

View full text Add to dashboard Cite

This paper presents a review on vigilance enhancement using computerized means and discusses their contradictory findings. It highlights the key differences between research findings and argues that variations in experimental protocol could be a significant contributing factor to the controversial results. In this paper, we found that, computerized means of enhancement are reliable and have significant effects on reducing vigilance decrement. The paper discusses the challenges toward the enhancement techniques and suggests several alternative strategies to reduce vigilance decrement. Furthermore, this review provides evidence to use computerized means of enhancement on vigilance studies, regardless of their practical challenges.

show abstract

Driver drowsiness detection based on non-intrusive metrics considering individual specifics

Cited by 138 publications

References 13 publications

Non-intrusive Driver Drowsiness Detection based on Face and Eye Tracking

Non-intrusive Driver Drowsiness Detection based on Face and Eye Tracking

Reliability of the Epworth Sleepiness Scale and the Berlin Questionnaire for screening obstructive sleep apnea syndrome in the context of the examination of candidates for drivers

Vigilance Enhancement Using Computerized Techniques

Contact Info

Product

Resources

About