Contactless Pulse Rate Assessment: Results and Insights for Application in Driving Simulators
DOI: 10.3390/app15179512
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study investigates the feasibility of using remote photoplethysmography (rPPG) for contactless pulse rate (PR) assessment in dynamic driving simulator environments. While wearable sensors are common, they suffer from motion artifacts and discomfort, whereas laboratory-grade contact sensors are impractical for driving scenarios. rPPG offers a non-invasive alternative by detecting subtle facial color changes via video, but its reliability in high-motion settings remains debated. The research specifically addresses whether video-based PR estimation is viable during simulated driving, if it can detect age-related physiological differences, and if Eulerian Video Magnification (EVM) significantly improves accuracy despite its computational cost. The methodology utilizes a retrospective dataset of 79 video recordings from 65 healthy participants across two age groups (30–45 and 60–75 years). Videos were captured using a low-cost RGB webcam, with reference PR data collected simultaneously via Empatica E4 wrist sensors. The processing pipeline involves face detection using YOLO v8, eye exclusion to isolate facial skin, and segmentation into overlapping 30-second sequences. The core analysis compares PR extraction with and without EVM. EVM was applied using Gaussian and Laplacian pyramids, temporal filtering (0.4–3 Hz), and a magnification factor of 20. Signal extraction involved calculating mean pixel intensity, applying Principal Component Analysis (PCA), and detecting peaks using a modified Pan–Tompkins algorithm. The study also evaluated the Empatica E4’s potential bias by comparing it against Faros 360 measurements in an independent dataset. Results indicate that rPPG is feasible for PR monitoring in driving simulators, though accuracy varies. Applying EVM slightly improved performance, reducing the mean absolute error (MAE) from 6.48 bpm to 5.04 bpm compared to the reference sensor. Under strict conditions, the lowest MAE achieved was approximately 2 bpm. However, EVM introduced a computational overhead of about 20 seconds for every 30-second video sequence. The study identified statistically significant differences in PR between younger and older drivers in both reference and rPPG data, confirming the method’s ability to detect demographic physiological variations. Additionally, the analysis highlighted that error rates increased with higher reference pulse rates, consistent with known limitations of wearable sensors during elevated activity. The findings demonstrate that contactless PR monitoring is a viable tool for driver state assessment, offering an unobtrusive alternative to wearables. While EVM provides marginal accuracy improvements, its time complexity suggests it may be redundant for real-time applications where slight errors are acceptable. The confirmation of age-related PR differences supports the use of rPPG for personalized monitoring systems. These results encourage further research into optimizing signal processing for dynamic environments, potentially integrating rPPG into in-car safety systems to detect stress or fatigue without requiring physical contact with the driver.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-07 |
| archive | success | openalex | — | — | 5 | 2026-06-09 |
| extract | success | cached | — | — | 2 | 2026-06-09 |
| clean | success | clean | — | — | 1 | 2026-06-07 |
| chunk | success | chunk | — | — | 1 | 2026-06-07 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-07 |
| promote | success | — | — | — | 1 | 2026-06-07 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-09 |
| tag | success | vector_similarity | — | — | 8 | 2026-06-11 |
| verify | success | — | — | — | 1 | 2026-06-09 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-09; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Empirical Findings: physiological data
- Methodological Resource: tool software