An Improvement for Video-based Heart Rate Variability Measurement

Peixi Li (1) , Yannick Benezeth (1) , Keisuke Nakamura (2) , Randy Gomez (2) , Chao Li (3) , Fan Yang (1)

1 ImViA - Imagerie et Vision Artificielle [Dijon]
2 Honda Research Institute
3 State Key Laboratory of Acoustics, Institute of Acoustics, Chinese Academy of Sciences

Remote photoplethysmography (RPPG) is a technique to measure the physiological signs (such as Heart Rate) remotely. Compared with the conventional photoplethysmography (PPG), the RPPG provides long term monitoring without contact equipment and makes the application possible and comfortable in daily life. Heart Rate Variability (HRV) is a medical index which is calculated as the interval of the heartbeats. It has been proved that the HRV can be used as a biomarker for the autonomic nervous system (ANS) so that the features of HRV are possible to be utilized to detect the human stress and other emotion states [1] [2]. To make the RPPG method works more effectively in human emotion states detection, the measurement of HRV should be improved. In this paper, we propose to use the slope sum function (SSF) to improve the interbeat interval (IBI) detection. The performance of this new method was evaluated with the HRV features which can be used for emotion detection. The results showed that this new method has improved the accuracy of the HRV measurement in both frequency domain and time domain.

Mots clés: autonomic nervous system; human stress; RPPG method; interbeat interval detection; HRV features; HRV measurement; human emotion state detection; Heart rate variability; video-based heart rate variability measurement; slope sum function; Frequency-domain analysis; frequency domain; Measurement uncertainty; time domain; Stress; heart rate variability (hrv); feature extraction; slope sum function (SSF); neurophysiology; emotion recognition; patient monitoring; remote photoplethysmography (RPPG); photoplethysmography; video signal processing; physiological signs; medical index

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Traitement du signal et de l'image [eess.SP]

Peixi Li, Yannick Benezeth, Keisuke Nakamura, Randy Gomez, Chao Li, et al.. An Improvement for Video-based Heart Rate Variability Measurement. 2019 IEEE 4th International Conference on Signal and Image Processing (ICSIP), Jul 2019, Wuxi, China. pp.435-439, ⟨10.1109/SIPROCESS.2019.8868712⟩. ⟨hal-02438311⟩