Automated non-contact heart rate measurement using conventional video cameras

In this work we propose an algorithm for noncontact heart rate estimation using conventional cameras. The algorithm is robust to movements and works under non-controlled illumination. We divide the frames in micro-regions that are tracked using an optical flow algorithm to compensate for movements and we apply a clustering algorithm to automatically select the best micro-regions to use for heart rate estimation. We also propose a temporal and spatial filtering scheme to reduce noise and an adaptive filter to improve the signal to noise ratio of the signal employed for heart rate estimation. We evaluated our algorithm for real and synthetic data, comparing its results to a fingertip pulse oximeter and a state-of-the-art algorithm for heart rate estimation through video. Results show that the algorithm can work well under challenging situations