Car-Driver Drowsiness Assessment through 1D Temporal Convolutional Networks

TL;DR

This paper presents a novel driver drowsiness detection system using a custom bio-sensor and a 1D Temporal Convolutional Network, achieving high accuracy in real-time classification to improve road safety.

Contribution

It introduces an innovative bio-sensor combined with a hyper-filtering technique and a 1D TCN architecture for accurate, real-time driver drowsiness detection.

Findings

Achieved approximately 96% accuracy in drowsiness classification.

Developed a new bio-sensor with near-infrared LEDs and photo-detectors.

Implemented an embedded hyper-filtering and dilation setup for real-time processing.

Abstract

Recently, the scientific progress of Advanced Driver Assistance System solutions (ADAS) has played a key role in enhancing the overall safety of driving. ADAS technology enables active control of vehicles to prevent potentially risky situations. An important aspect that researchers have focused on is the analysis of the driver attention level, as recent reports confirmed a rising number of accidents caused by drowsiness or lack of attentiveness. To address this issue, various studies have suggested monitoring the driver physiological state, as there exists a well-established connection between the Autonomic Nervous System (ANS) and the level of attention. For our study, we designed an innovative bio-sensor comprising near-infrared LED emitters and photo-detectors, specifically a Silicon PhotoMultiplier device. This allowed us to assess the driver physiological status by analyzing the associated PhotoPlethysmography (PPG) signal.Furthermore, we developed an embedded time-domain hyper-filtering technique in conjunction with a 1D Temporal Convolutional architecture that embdes a progressive dilation setup. This integrated system enables near real-time classification of driver drowsiness, yielding remarkable accuracy levels of approximately 96%.