A Portable Fuzzy Driver Drowsiness Estimation System

Abstract

The adequate automatic detection of driver fatigue is a very valuable approach for the prevention of traffic accidents. Devices that can determine drowsiness conditions accurately must inherently be portable, adaptable to different vehicles and drivers, and robust to conditions such as illumination changes or visual occlusion. With the advent of a new generation of computationally powerful embedded systems such as the Raspberry Pi, a new category of real-time and low-cost portable drowsiness detection systems could become standard tools. Usually, the proposed solutions using this platform are limited to the definition of thresholds for some defined drowsiness indicator or the application of computationally expensive classification models that limits their use in real-time. In this research, we propose the development of a new portable, low-cost, accurate, and robust drowsiness recognition device. The proposed device combines complementary drowsiness measures derived from a temporal window of eyes (PERCLOS, ECD) and mouth (AOT) states through a fuzzy inference system deployed in a Raspberry Pi with the capability of real-time response. The system provides three degrees of drowsiness (Low-Normal State, Medium-Drowsy State, and High-Severe Drowsiness State), and was assessed in terms of its computational performance and efficiency, resulting in a significant accuracy of 95.5% in state recognition that demonstrates the feasibility of the approach.

A Review of Recent Developments in Driver Drowsiness Detection Systems

“…In terms of detection accuracy, the DDD imagebased systems differ in their results. Since they monitor features that are highly correlated to drowsiness, such as yawning, blinking, head movement, and eye closure, most of them have achieved high accuracy, between 85% to 99%, as shown in systems [17,52,54,55,59]. However, it should be noted that such systems are affected by multiple factors, as mentioned previously in the challenges section, and are often implemented and tested in a controlled environment or using existing DDD video datasets [30,33,49,59,63,68].…”

“…Other issues associated with such systems are the presence of additional features on the face, such as sunglasses, a beard, or a mustache, that may cover the eye or mouth and lead to a system failure. Additional challenges include the random head movement [38,41,52], different skin colors, various lighting conditions [55,130], face's distance from the camera, different face structure based on race, and real-time video analysis that require powerful computing resources [103]. All of that may reduce the accuracy or even lead to false detection.…”

“…Eye and mouth analysis Celecia et al [52] proposed a low-cost, portable, robust, and accurate DDD device that used an infrared illuminator and camera to record images. The device's processing model, which was performed over a Raspberry Pi 3 Model B, combines features obtained from the eyes and mouths of the subjects under consideration.…”

Trends and Future Prospects of the Drowsiness Detection and Estimation Technology

“…In this case, the classification of “wakefulness”, “fatigue”, and “dozing” was performed, and the average classification accuracy was about 65%. There are other previous studies that use PERCLOS to detect or estimate drowsiness [ 62 ].…”

“…Celecia et al’s case study combined measures of sleepiness based on information about the eyes (PERCLOS, eye closure time) and mouth free time via a fuzzy inference system on a Raspberry Pi system to enable real-time responses [ 70 ]. The system detected three drowsiness levels (low/normal, medium/drowsy, and high/severe drowsiness), which were evaluated in terms of their computational performance and efficiency, resulting in a significant accuracy of 95.5% in state recognition [ 62 ]. Dua et al used four deep learning models, AlexNet, VGG-FaceNet, FlowImageNet, and ResNet, to detect sleepiness by considering four different types of features: hand gestures, facial expressions, behavioral features, and head movements.…”