Abstract
BackgroundRemote photoplethysmography (rPPG) is a promising optical method for non-contact assessment of pulse rate (PR) from video recordings. In order to implement the method in real-time applications, it is necessary for the rPPG algorithms to be capable of eliminating as many distortions from the pulse signal as possible.MethodsIn order to increase the degrees-of-freedom of the distortion elimination, the dimensionality of the RGB video signals is increased by the wavelet transform decomposition using the generalized Morse wavelet. The proposed Continuous-Wavelet-Transform-based Sub-Band rPPG method (SB-CWT) is evaluated on the 101 publicly available RGB facial video recordings and corresponding reference blood volume pulse (BVP) signals taken from the MMSE-HR database. The performance of the SB-CWT is compared with the performance of the state-of-the-art Sub-band rPPG (SB).ResultsMedian signal-to-noise ratio (SNR) for the proposed SB-CWT ranges from 6.63 to 10.39 dB and for the SB from 4.23 to 6.24 dB. The agreement between the estimated PRs from rPPG pulse signals and the reference signals in terms of the coefficients of determination ranges from 0.81 to 0.91 for SB-CWT and from 0.41 to 0.47 for SB. All the correlation coefficients are statistically significant (p < 0.001). The Bland–Altman plots show that mean difference range from 5.37 to 1.82 BPM for SB-CWT and from 22.18 to 18.80 BPM for SB.DiscussionThe results show that the proposed SB-CWT outperforms SB in terms of SNR and the agreement between the estimated PRs from RGB video signals and PRs from the reference BVP signals.
Highlights
Remote photoplethysmography is a non-contact optical method that measures the intensity of the light reflected from the skin by the means of a digital camera
There are numerous datasets that are in general suitable for the evaluation of the Remote photoplethysmography (rPPG) algorithms, but to the best of our knowledge MMSE-HR is the only one that offers uncompressed facial recordings. This type of recordings is needed for the evaluation of the methods that exploit model-based algorithms for rPPG signal extraction
At l = 32, l = 64 and l = 128, Sub-band rPPG (SB)-Continuous Wavelet Transform (CWT) segments the input rPPG signal on more sub-bands, which in turn increases the effectiveness of the separation of the pulse signal from noise components
Summary
Remote photoplethysmography (rPPG) is a non-contact optical method that measures the intensity of the light reflected from the skin by the means of a digital camera. In the end-diastole phase the transmural pressure is the lowest and the amount of back-reflected light is maximal In this model the periodic changes of light scattering and absorption due to the deformation of the connective tissue of the skin are presumed to form the rPPG signal (Kamshilin et al, 2015). RPPG signal, just like the conventional contact photoplethysmogram (PPG), consists of the stationary (DC) and varying (AC) part The former is related to the skin structure and the average blood volume in arterial and venous system (Tamura et al, 2014). The proposed Continuous-Wavelet-Transform-based Sub-Band rPPG method (SB-CWT) is evaluated on the 101 publicly available RGB facial video recordings and corresponding reference blood volume pulse (BVP) signals taken from the MMSE-HR database. Discussion: The results show that the proposed SB-CWT outperforms SB in terms of SNR and the agreement between the estimated PRs from RGB video signals and PRs from the reference BVP signals
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
