Non-Invasive Driver Drowsiness Detection System

Abstract

Drowsiness when in command of a vehicle leads to a decline in cognitive performance that affects driver behavior, potentially causing accidents. Drowsiness-related road accidents lead to severe trauma, economic consequences, impact on others, physical injury and/or even death. Real-time and accurate driver drowsiness detection and warnings systems are necessary schemes to reduce tiredness-related driving accident rates. The research presented here aims at the classification of drowsy and non-drowsy driver states based on respiration rate detection by non-invasive, non-touch, impulsive radio ultra-wideband (IR-UWB) radar. Chest movements of 40 subjects were acquired for 5 m using a lab-placed IR-UWB radar system, and respiration per minute was extracted from the resulting signals. A structured dataset was obtained comprising respiration per minute, age and label (drowsy/non-drowsy). Different machine learning models, namely, Support Vector Machine, Decision Tree, Logistic regression, Gradient Boosting Machine, Extra Tree Classifier and Multilayer Perceptron were trained on the dataset, amongst which the Support Vector Machine shows the best accuracy of 87%. This research provides a ground truth for verification and assessment of UWB to be used effectively for driver drowsiness detection based on respiration.

Respiration Based Non-Invasive Approach for Emotion Recognition Using Impulse Radio Ultra Wide Band Radar and Machine Learning

“…The procedure was replicated several times against various participants. The obtained RPM is validated against a commercial Pulse oximeter used in [44] and shown in Figure 11.…”

“…Because of the IR-UWB radar's extremely low emission power, it has no adverse effect on the human body and passes the safety standards. Because of its non-ionizing, non-intrusive, non-tackling, no blind spot angles, and ability to penetrate a range of materials or barriers, the IR-UWB radar offers benefits over other existing tools for recording the physiological signals [42][43][44].…”

“…Using the Fourier transform, the frequency spectrum of the gathered signal is obtained. The maximum frequency of an adult's respiration rate is 0.4 Hz [44,[51][52][53][54]. To get the respiration signal, a filter with a cut-off frequency of 0.4 Hz is required.…”

Wrong-Lane Accidents Detection using Random Forest Algorithm in comparison with Decision Tree for Improved Accuracy

“…Based on the above summary, if overall prediction performance is the primary concern, then the integrated method should be selected, in which the RF models with only a few significant variables identified by DT or important variables identified by the tree , can be entered to achieve more precision(Gazder, Ahmed, and Shahid 2021). If the focus is on major accident prediction performance, the integrated method, where RF models with only a few significant variables identified by DT, or major variables identified by the Random Forest, or DT models with only a few significant variables identified by DT should be selected for greater sensitivity (Siddiqui et al 2021). The accuracy of the DT classification algorithm depends on the size of the training and testing data set.…”

A survey on driver drowsiness detection using physiological, vehicular, and behavioral approaches

“…The absence of sharpness, produced by the oblivious change from a state of alertness to a state of rest, prompts several mishaps [1]. The U.S. Public highway traffic safety administration (NHTSA) reports that sluggish driving brought about very nearly 100,000 mishaps and more than 1,500 passings every year [2]. A driver's weariness can stem from a number of causes, including lack of rest, stress, anxiety, and alcohol consumption.…”