Title: Non-contact heart rate measurement from face video sequences using SIFT and FastICA

Authors: Hemlata G. Biradar; Jayanand P. Gawande

Addresses: Department of Electronics Engineering, Ramrao Adik Institute of Technology, D.Y. Patil (Deemed to be University), Navi Mumbai, India ' Department of Instrumentation Engineering, Ramrao Adik Institute of Technology, D.Y. Patil (Deemed to be University), Navi Mumbai, India

Abstract: Traditional techniques of heart rate (HR) measurement rely upon optical or electronic sensors. This research proposes a HR measuring method in a non-contact way that enables a cardiac pulse's physiological examination without using electrodes. The proposed method is based on automated face tracking and blind source separation of the colour channels into separate components from colour video recordings of the human face using fast independent component analysis (FastICA). To reduce noise caused by face motion, we used the scale-invariant feature transform (SIFT) and affine transform (AT). The experimentation is carried out on COHFACE and DROZY datasets. The HR predicted by the proposed method and by the finger blood volume pulse sensor is compared using the Bland-Altman and correlation analysis. The proposed method when compared to other methods utilising comparable datasets outperforms in terms of mean absolute error (MSE), root mean square error (RMSE), standard deviation (SD), and correlation coefficient (CC).

Keywords: heart rate; HR; scale invariant feature transform; SIFT; fast independent component analysis; FastICA; photoplethysmography; PPG; power spectrum.

DOI: 10.1504/IJBET.2024.138617

International Journal of Biomedical Engineering and Technology, 2024 Vol.45 No.1, pp.65 - 81

Received: 03 Apr 2023
Accepted: 25 Aug 2023
Published online: 16 May 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article