Detecting of Fatigue States of a Car Driver

Bittner, Roman; Smrčka, Pavel; Vysoký, Petr; Hána, Karel; Pousek, L.; Schreib, Petr

doi:10.1007/3-540-39949-6_32

Cited by 38 publications

(27 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the prediction of the system when tested with unknown vectors gave the average drowsiness level of 3. Hence, it is not able to predict extreme cases with the available rule base, mostly outputting scale 3 from the [1][2][3][4][5] interval indicating a moderate level of drowsiness (Fig. 7).…”

Section: Comparison Of Prediction Performancesmentioning

confidence: 99%

“…It may output false classification results in some extreme cases. It is able to follow the drowsiness level defined by average of experimenter's, independent assessors', and subject's own drowsiness gradings on the [1][2][3][4][5] scale. An S-T FIS system based on the F 1 (visual clues) feature space gives reasonably good results when 18 completely unknown sessions were used as input (Fig.…”

Section: Comparison Of Prediction Performancesmentioning

confidence: 99%

“…10). The results are also filtered using heuristics to damp the overshoots of the system and to rescale it into a [1][2][3][4][5] scale.…”

Section: Comparison Of Prediction Performancesmentioning

confidence: 99%

“…The aim for monitoring is to provide a solution to the drowsiness problem on roads from an engineering perspective. Several studies have been conducted in order to explain the phenomenon or to produce a reliable measure for on-road driver monitoring technologies [1][2][3][4][5][6]. One pragmatic study produced a successful metric known as the percentage closure of the eyelid (PERCLOS), obtainable by computer vision (CV) techniques [7].…”

Section: Introductionmentioning

confidence: 99%

“…For this reason, various researchers have conducted studies to combine and correlate the signals to overcome the deficiency of single-input systems. Bittner et al [2] correlated the symptoms from three different signals, namely steering-wheel adjustments, vertical electro-oculography for eye movements, and several parameters by independent assessors obtained by watching the video of the driving simulations. In a similar study, Brookhuis and De Waard [3] correlated physiological signals such as electroencephalography with vehicle-related performance measures, namely car-following performance and number of steering-wheel reversals.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Multi-sensor driver drowsiness monitoring

Boyraz

Acar

Kerr

2008

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

A system for driver drowsiness monitoring is proposed, using multi-sensor data acquisition and investigating two decision-making algorithms, namely a fuzzy inference system (FIS) and an artificial neural network (ANN), to predict the drowsiness level of the driver. Drowsiness indicator signals are selected allowing non-intrusive measurements. The experimental set-up of a driver-drowsiness-monitoring system is designed on the basis of the soughtafter indicator signals. These selected signals are the eye closure via pupil area measurement, gaze vector and head motion acquired by a monocular computer vision system, steering wheel angle, vehicle speed, and force applied to the steering wheel by the driver. It is believed that, by fusing these signals, driver drowsiness can be detected and drowsiness level can be predicted. For validation of this hypothesis, 30 subjects, in normal and sleep-deprived conditions, are involved in a standard highway simulation for 1.5 h, giving a data set of 30 pairs. For designing a feature space to be used in decision making, several metrics are derived using histograms and entropies of the signals. An FIS and an ANN are used for decision making on the drowsiness level. To construct the rule base of the FIS, two different methods are employed and compared in terms of performance: first, linguistic rules from experimental studies in literature and, second, mathematically extracted rules by fuzzy subtractive clustering. The drowsiness levels belonging to each session are determined by the participants before and after the experiment, and videos of their faces are assessed to obtain the ground truth output for training the systems. The FIS is able to predict correctly 98 per cent of determined drowsiness states (training set) and 89 per cent of previously unknown test set states, while the ANN has a correct classification rate of 90 per cent for the test data. No significant difference is observed between the FIS and the ANN; however, the FIS might be considered better since the rule base can be improved on the basis of new observations.

show abstract

Section: Comparison Of Prediction Performancesmentioning

confidence: 99%

Section: Comparison Of Prediction Performancesmentioning

confidence: 99%

“…10). The results are also filtered using heuristics to damp the overshoots of the system and to rescale it into a [1][2][3][4][5] scale.…”

Section: Comparison Of Prediction Performancesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Multi-sensor driver drowsiness monitoring

Boyraz

Acar

Kerr

2008

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

show abstract

Fatigue

Roca-Dorda

González

Martínez-González

et al. 2006

Wiley Encyclopedia of Biomedical Engineering

View full text Add to dashboard Cite

The study of fatigue has increased by leaps and bounds; however, still no universal definition exists and fatigue remains a vague concept used as an umbrella term that comprises of physiological, emotional, and behavioral factors that can result in chronic physical or mental states affecting the cognitive, memory, motoric, and perceptive capacities. Biomedical engineering has faced fatigue evaluation from two different approaches: exercise‐induced fatigue (or metabolic fatigue) and central fatigue (related to the evolution of the subject condition caused by the execution of normal tasks). As metabolic fatigue is a complex mechanism usually discussed by exercise physiology, this work will center on central fatigue detection and evaluation, because of its importance within the field of human factors engineering in occupational health studies. Central fatigue has been referred to in various terms that are more or less mutually interchangeable in the research literature: loss of alertness or vigilance, fatigue, and drowsiness, although no significant difference exists among them. Research has focused on the development of objective and reliable measurement techniques able to quantify the state of human deteriorated condition, although none of the proposed methods has been yet considered to be a gold standard. This work reviews several fatigue indicators that have been reported to reflect the different dimensions resulting from this state [heart‐related measures such as the heart rate (HR) and heart rate variability (HRV), muscles activity through electromyogram (EMG) and mechanomyogram (MMG), spectral analysis of the electroencephalograph (EEG) and event‐related potential (ERP), ocular metrics, speech parameters, etc].

show abstract