The principle of detection is based on the fact that the human skin becomes more red when the blood pressure is at its maximum (systolic pressure), and less red when the pressure is at its minimum (diastolic pressure). For people not having arrhythmia, these changes are periodic, and therefore, its’ frequency can be caught and measured using spectrum analysis principles.
Unlike the method in the publication above, which analyses object motions in the series of video frames, DataArt’s research application uses a simpler, yet effective method of tracking the change in Red component of a set of pixels in time. These data series are spectrally analyzed then, yielding a frequency response curve each. The obtained curves, in turn, and validated to filter out those containing only noise, or off-range spectrum peaks. The remaining curves are statistically processed to determine a frequency which is most probably corresponds to the user’s heartbeat rate.
The experiments have shown that with the use of a 2Mpix high-grade web camera, the method is capable of detecting and calculating the user’s heartbeat rate with the camera-to-face distance up to as long as up 2ft. Using less quality camera, like Kinect built-it 640*480 one, shortens the distance, yet allowing for remote measurement.