Multi-sensor Driver Drowsiness Monitoring

Journal article, 2008

A system for driver drowsiness monitoring is proposed, using multi-sensor dataacquisition 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 experi-mental set-up of a driver-drowsiness-monitoring system is designed on the basis of the sought-after 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 wheelangle, vehicle speed, and force applied to the steering wheel by the driver. It is believed that, byfusing 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, areinvolved in a standard highway simulation for 1.5 h, giving a data set of 30 pairs. For designing afeature space to be used in decision making, several metrics are derived using histograms andentropies of the signals. An FIS and an ANN are used for decision making on the drowsinesslevel. To construct the rule base of the FIS, two different methods are employed and comparedin terms of performance: first, linguistic rules from experimental studies in literature and,second, mathematically extracted rules by fuzzy subtractive clustering. The drowsiness levelsbelonging 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 thesystems. 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 correctclassification rate of 90 per cent for the test data. No significant difference is observed betweenthe FIS and the ANN; however, the FIS might be considered better since the rule base can beimproved on the basis of new observations.