| Title |
Assessment of Heart Rate Derivation Methods and Applicability in Remote Photoplethysmography |
| Authors |
채종의(JongEui Chae) ; 김대열(DaeYeol Kim) ; 이광기(KwangKee Lee) ; 박찬형(ChanHyeong Park) |
| DOI |
https://doi.org/10.5573/ieie.2023.60.10.35 |
| Keywords |
Remote photoplethysmography; Fair evaluation; FFT; Peak detection |
| Abstract |
In this study, we compare and analyze the commonly used frequency domain analysis and time domain analysis methods, which are representative approaches for deriving heart rate (HR) from the blood volume pulse (BVP) in remote photoplethysmography (rPPG) technology. We aim to present the advantages, necessity, and validity of the Peak Detection-based HR measurement method. Typically, in rPPG, the HR is often measured using frequency domain analysis by applying a band-pass filter (BPF) for post-processing and performing FFT on the signal. However, this approach makes the shape analysis of the signal challenging and reduces its applicability. Furthermore, we identified that the frequency domain analysis method alters the principal component of BVP, leading to limitations in HR calculation. To overcome this limitation, we propose a Peak Detection-based HR measurement method. Through experimentation, we compared the frequency analysis method with the Peak Detection-based HR assessment metric over different measurement time intervals. The results demonstrate that Peak Detection offers greater validity as a method for analyzing physiological signals compared to the frequency analysis approach. |