A Comparison of Algorithms to Estimate Heart and Respiratory Rate from Face Video Processing

The growing importance of remote monitoring of heart rate (HR) and respiratory rate (RR) has heightened the relevance of noncontact measurement methods. In this chapter, we compare the performance of independent component analysis (ICA) method and Green–Red Differences (GRD) method in extracting the remote photoplethysmography (rPPG) signal from both simulated and real images captured by a webcam and compute the instantaneous values of HR and RR using the processed signals. In all cases, the best algorithm was GRD with Fourier analysis, so this method should be analyzed in greater depth to propose improvements in its performance and make it a reliable method for the estimation of HR and RR.