A Survey on Drowsiness Detection -- Modern Applications and Methods

TL;DR

This paper addresses drowsiness detection as a safety-critical problem across industries beyond driving. It classifies methods into three measuring modalities: $EEG$, $ECG$, and vision-based sensing, and reviews modern algorithms, datasets, and evaluation metrics. It identifies weaknesses such as real-time constraints, data transmission stability, and bias, and proposes practical remedies including synthetic data, model compression, and multimodal fusion. Overall, the work provides a comprehensive, forward-looking view of drowsiness detection with broad opportunities for future research and deployment.

Abstract

Drowsiness detection holds paramount importance in ensuring safety in workplaces or behind the wheel, enhancing productivity, and healthcare across diverse domains. Therefore accurate and real-time drowsiness detection plays a critical role in preventing accidents, enhancing safety, and ultimately saving lives across various sectors and scenarios. This comprehensive review explores the significance of drowsiness detection in various areas of application, transcending the conventional focus solely on driver drowsiness detection. We delve into the current methodologies, challenges, and technological advancements in drowsiness detection schemes, considering diverse contexts such as public transportation, healthcare, workplace safety, and beyond. By examining the multifaceted implications of drowsiness, this work contributes to a holistic understanding of its impact and the crucial role of accurate and real-time detection techniques in enhancing safety and performance. We identified weaknesses in current algorithms and limitations in existing research such as accurate and real-time detection, stable data transmission, and building bias-free systems. Our survey frames existing works and leads to practical recommendations like mitigating the bias issue by using synthetic data, overcoming the hardware limitations with model compression, and leveraging fusion to boost model performance. This is a pioneering work to survey the topic of drowsiness detection in such an entirely and not only focusing on one single aspect. We consider the topic of drowsiness detection as a dynamic and evolving field, presenting numerous opportunities for further exploration.

Figures (2)

• Figure 1: Depicts areas of applications in urgent need of accurate and real-time drowsiness detection task.
• Figure 2: Overall structure of the surveyed advancements in drowsiness detection. This primary framework illustrates the three measuring techniques utilized in drowsiness detection, along with their respective attributes and methodologies. From the considered topology, we first distinguish between drowsiness detection based on physiological signals and monitoring through vision. Based on these classification, we assign different attributes to individual measuring techniques. The sensing modalities for EEG and ECG are not mutually exclusive and can be divided under wired and wireless application. While under the vision-approach, we now consider static investigation for single frame and dynamic investigation for multiple frames and the unobtrusiveness for the general setup. Abbreviations like HRV stands for heartrate variability, PRV for pulserate variability, QRS for morphological components of a heartbeat, LF/HF for certain frequency bands, and finally PERCLOS for percentage of eye closure. The sensing modality demonstrates the potential realization of applications. The bottom row outlines the typical areas of application for drowsiness detection across all measuring techniques.