Thermal Imaging for Contactless Cardiorespiratory and Sudomotor Response Monitoring

TL;DR

This paper characterizes the extraction of these three biosignals from facial thermal video using a signal-processing pipeline that tracks anatomical regions, applies spatial aggregation, and separates slow sudomotor trends from faster cardiorespiratory components.

Abstract

Thermal infrared imaging captures skin temperature changes driven by autonomic regulation and can potentially provide contactless estimation of electrodermal activity (EDA), heart rate (HR), and breathing rate (BR). While visible-light methods address HR and BR, they cannot access EDA, a standard marker of sympathetic activation. This paper characterizes the extraction of these three biosignals from facial thermal video using a signal-processing pipeline that tracks anatomical regions, applies spatial aggregation, and separates slow sudomotor trends from faster cardiorespiratory components. For HR, we apply an orthogonal matrix image transformation (OMIT) decomposition across multiple facial regions of interest (ROIs), and for BR we average nasal and cheek signals before spectral peak detection. We evaluate 288 EDA configurations and the HR/BR pipeline on 31 sessions from the public SIMULATOR STUDY 1 (SIM1) driver monitoring dataset. The best fixed EDA configuration (nose region, exponential moving average) reaches a mean absolute correlation of $0.40 \pm 0.23$ against palm EDA, with individual sessions reaching 0.89. BR estimation achieves a mean absolute error of $3.1 \pm 1.1$ bpm, while HR estimation yields $13.8 \pm 7.5$ bpm MAE, limited by the low camera frame rate (7.5 Hz). We report signal polarity alternation across sessions, short thermodynamic latency for well-tracked signals, and condition-dependent and demographic effects on extraction quality. These results provide baseline performance bounds and design guidance for thermal contactless biosignal estimation.

Figures (6)

• Figure 1: Thermal facial video can be converted into ROI temperature traces and decomposed into respiratory, cardiac-related, and slow autonomic components, including EDA-like trends and perinasal perspiration proxies.
• Figure 2: Pipeline overview. Thermal frames are processed by a face detector to localise landmarks. Six facial ROIs are defined and scalar temperature traces are extracted per frame. Temporal decomposition separates slow sudomotor trends, cardiac pulse, and respiratory components for comparison with synchronised contact ground truth.
• Figure 3: $\text{PCC}_{\text{abs}}$ against PEDA for all 48 ROI-method combinations ($n = 31$).
• Figure 4: T003-CD: thermal EDA trends from three ROIs vs PEDA ground truth (black).
• Figure 5: $\text{PCC}_{\text{abs}}$ against PEDA by driving condition ($n = 7$--$8$ per task).