Equitable non-contact infrared thermography after solar loading using deep learning

TL;DR

This paper introduces SL-Net, a deep learning model that corrects solar loading effects in infrared thermography, improving fever detection accuracy and fairness across skin tones without lengthy reacclimation.

Contribution

The study presents the first analysis of solar loading impact on IRT and develops a novel deep learning approach to mitigate this effect, enhancing robustness and equity in thermography.

Findings

SL-Net reduces solar loading error by 68%.

Forehead skin temperature increases by 2°C after solar loading.

SL-Net is unbiased across skin tones.

Abstract

Widely deployed for fever detection, infrared thermometers (IRTs) enable rapid non-contact measurement of core body temperature but are inaccurate in unconstrained environments when skin temperature is transient. In this work, we present the first study on the effect of solar loading--solar radiation-induced elevation of skin but not core temperature--on IRT performance. Solar loading causes poor specificity in IRT fever detection, and the standard procedure is to reacclimate subjects for up to 30 minutes before IRT measurement. In contrast, we propose a single-shot deep learning model that removes solar loading transients from thermal facial images, allowing accurate IRT operation in solar loaded conditions. Forehead skin temperature increases by 2.00{\deg}C after solar loading, and our deep learning model, SL-Net, reduces this error by 68\% to 0.64{\deg}C. We show that the solar loading effect depends on skin tone, introducing inequity in IRT performance, while SL-Net is unbiased. We open source a diverse dataset of 100 subjects with co-registered RGB-thermal images, and IRT and skin tone measurements. Our work shows that it is possible to use machine learning to correct complex thermal perturbations to enable robust and equitable human thermography.