Abstract

This study investigates the potential of using multispectral imaging for detecting physical stress on human beings. Participants were recruited to obtain multispectral images, and a proposed facial muscle activity detection algorithm was established without background information. The algorithm model is verified with respect to physical stress ground truth in order to classify the baseline and physical stress status. The algorithm achieved better results in the experiment with an accuracy rate of 75%, which will provide a foundation for future industrialization. Experimental results demonstrated that multispectral imaging, as a non-invasive method, has the potential to identify physical stress on humans.

© 2018 Optical Society of America

Heart rate estimation from facial videos with motion interference using T-SNE-based signal separation

Hequn Wang, Xuezhi Yang, Xuenan Liu, and Dingliang Wang
Biomed. Opt. Express 13(9) 4494-4509 (2022)

Assessment of mental stress effects on prefrontal cortical activities using canonical correlation analysis: an fNIRS-EEG study

Fares Al-Shargie, Tong Boon Tang, and Masashi Kiguchi
Biomed. Opt. Express 8(5) 2583-2598 (2017)

In vivo detection of exercise-induced ultrastructural changes in genetically-altered murine skeletal muscle using polarization-sensitive optical coherence tomography

James J. Pasquesi, Simon C. Schlachter, Marni D. Boppart, Eric Chaney, Stephen J. Kaufman, and Stephen A. Boppart
Opt. Express 14(4) 1547-1556 (2006)

Multispectral opto-acoustic tomography of exercised muscle oxygenation

Gael Diot, Alexander Dima, and Vasilis Ntziachristos
Opt. Lett. 40(7) 1496-1499 (2015)

Three-dimensional facial digitization using advanced digital image correlation

Hieu Nguyen, Hien Kieu, Zhaoyang Wang, and Hanh N. D. Le
Appl. Opt. 57(9) 2188-2196 (2018)