@MISC\{IMM2014-06796,
    author       = "J. N. Jensen and M. Hannemose",
    title        = "Camera-based Heart Rate Monitoring",
    year         = "2014",
    publisher    = "Technical University of Denmark, Department of Applied Mathematics and Computer Science",
    address      = "Richard Petersens Plads, Building 324, {DK-}2800 Kgs. Lyngby, Denmark, compute@compute.dtu.dk",
    note         = "{DTU} supervisor: Rasmus Larsen, rlar@dtu.dk, {DTU} Compute, and Knut Conradsen, knco@dtu.dk, {DTU} Compute",
    url          = "http://www.compute.dtu.dk/English.aspx",
    abstract     = "In recent years, measuring the heart rate using only a cheap camera has become popular. In this thesis we have implemented and tested different non-contact methods for measuring the heart rate. Face detection was used in all of these and {ICA} for separation of the observed signals into a heart rate signal was used in some. Videos of test subjects with a duration of 60 seconds were recorded. The methods were applied to these and were compared with a reference heart rate monitor.

The results were comparable or better than similar studies, and we achieved high correlation and small errors compared with the reference heart rate. We also found that the choice of colour space did not significantly change the results.

We have extended this method to calculate maps of how visible the heart rate is in different parts of the face, and used these to obtain better heart rate signals. Using these improved signals we have calculated the time delay of the heart rate between the hand and the face. We have compared this time difference with the systolic blood pressure, but we were unable to conclude whether there is a significant correlation, due to a too small sample size."
}