Heatmap-Based Method for Estimating Drivers' Cognitive Distraction
URL: http://arxiv.org/abs/2005.14136v2
archive: archived pipeline: cataloged verified
Abstract
In order to increase road safety, among the visual and manual distractions, modern intelligent vehicles need also to detect cognitive distracted driving (i.e., the drivers mind wandering). In this study, the influence of cognitive processes on the drivers gaze behavior is explored. A novel image-based representation of the driver's eye-gaze dispersion is proposed to estimate cognitive distraction. Data are collected on open highway roads, with a tailored protocol to create cognitive distraction. The visual difference of created shapes shows that a driver explores a wider area in neutral driving compared to distracted driving. Thus, support vector machine (SVM)-based classifiers are trained, and 85.2% of accuracy is achieved for a two-class problem, even with a small dataset. Thus, the proposed method has the discriminative power to recognize cognitive distraction using gaze information. Finally, this work details how this image-based representation could be useful for other cases of distracted driving detection.
Summary
Open-road on-highway study (Musabini and Chetitah, Valeo InnoCoRe) proposing a novel image-based heatmap representation of the driver's eye-gaze dispersion to detect cognitive distraction (mind wandering while still looking forward). Data were collected with a tailored protocol designed to induce cognitive distraction; gaze heatmaps were rendered onto a virtual wall and used as input to SVM classifiers. Drivers explored a wider visual area in neutral driving than under cognitive distraction. Two-class classification reached 85.2% accuracy on a small dataset.
Key finding
Gaze-dispersion heatmaps fed to SVMs detect cognitive distraction at 85.2% accuracy and capture the gaze-narrowing signature of mind wandering that pure eye-on-road detectors miss, addressing the Euro NCAP 2022 driver-inattentiveness requirement.
Methodology
Exp 1: 10 participants, repeated measures across 6 sessions from 26 total. Exp 2: 20 participants, Old/New sequence comparison. On-road driving paradigm with DRT and NASA-TLX measures.
Sample size: Exp 1: N=10; Exp 2: N=20
Quality score: 5 / 5