Heat stroke occurs due to an increase in core body temperature, making its monitoring crucial for prevention. Core body temperature is typically measured by inserting a probe into the body, but this method is invasive and impractical during physical activity. While wearable sensor-based estimation methods have been proposed, challenges remain regarding sensor wearability and cost.
To address these issues, this study explores a non-contact and cost-effective approach to estimating core body temperature using thermography. We assume a scenario where a subject is positioned in front of a thermographic camera in a hot environment, capturing thermal images of the subject’s face during exercise. To improve estimation accuracy, we extract surface temperatures from different facial regions in the acquired thermal images and construct a core body temperature estimation model using machine learning.
Published Paper:
- 平野 華織, 工藤 寛樹, 内山 彰, 桒野 聡, 千田 泰史, 丸谷 賢弘, 長谷川 凌佑, 横山 光樹, 中田 研.「サーモグラフィを用いた暑熱環境における非接触な深部体温推定法の検討」第86回全国大会講演論文集, vol. 2024, no. 1, pp. 153–154, 情報処理学会, 2024年3月
- 平野 華織, 工藤 寛樹, 内山 彰, 桑野 聡, 千田 泰史, 丸谷 賢弘, 長谷川 凌佑, 横山 光樹, 中田 研.「サーモグラフィを用いた顔の部位別表面温度に基づく深部体温推定法の検討」マルチメディア，分散，協調とモバイルシンポジウム2024論文集, vol. 2024, pp. 1297–1303, 情報処理学会, 2024年6月.
