Non-contact Heart Rate Measurement with Optimization of Variational Modal Decomposition Algorithm

In order to improve the accuracy of non-contact heart rate measurement, this study applied an approach by optimizing variational modal decomposition algorithm. Pulse wave signals were obtained by a plane-orthogonal-to-skin algorithm, and were further processed using an optimized variational modal decomposition (VMD). The optimized variational modal decomposition further optimizes the signal, automatically selecting the correct heart rate mode and eliminating noise. This method was further verified using UBFCRPPG (stands for Univ. Bourgogne Franche-Comté Remote PhotoPlethysmoGraphy) in a public dataset. Results showed that the optimized VMD can accurately measure the heart rates that meet the relevant standards of medical physiology. This study may provide an advanced approach for non-contact heart rate detection and will play a role in daily physiological measurement.