機械学習による顔面熱画像からの部位検出

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2021/04/01

タイトル(英語): Detecting Area from Facial Thermal Image by Machine Learning

著者名: 梅本　拓登（香川大学創造工学部），淺野　裕俊（香川大学創造工学部）

著者名(英語): Takuto Umemoto (Graduate School of Engineering, Kagawa University), Hirotoshi Asano (Graduate School of Engineering, Kagawa University)

キーワード: 一過性覚醒度，鼻部皮膚温度，カスケード分類器，YOLO　　transient arousal level，nose skin temperature，cascade classifier，YOLO

要約(英語): Temporary loss of attention due to lack of sleep is one of the leading causes of illness and traffic accidents. One of the researches to prevent them is detection technology of arousal level using the skin temperature of the nose, and a thermo camera is a non-invasive device for acquiring this. However, there are some conditions and problems in detecting a specific area directly from an infrared image. In this study, a model that can detect the face and nose regions from the thermal image corresponding to the displacement of the face by machine learning was created and evaluated aiming at application to the awakening level estimation technology. The results show that the model created may be useful.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.141 No.4 （2021） 特集：量子・情報・エレクトロニクス医療インタフェース

本誌掲載ページ: 582-583 p

原稿種別: 研究開発レター／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejeiss/141/4/141_582/_article/-char/ja/