0

I am working in research where we are using smart-phone camera to monitor users heart-rate using color variation as signal. What I did is getting the red color channel every 0.1 second (10Hz). The problem is that I am trying to use an FFT to get different frequencies that exist in the extracted signal and I used this Java code where the FFT function get as input two arrays (one for real part and one for img part of complex numbers). I saw also from this post that I can compute frequencies from the FFT function's results by using the formula:

freq = i * Fs / N

where Fs is the sampling rate and N is the number of points(input). The problem is that my sampling rate Fs, is too low (10Hz) and if I use above formula I am getting very low frquencies. Is there any other way to get frequencies?

Community
  • 1
  • 1
Milorenus Lomaliza
  • 221
  • 4
  • 8
  • 3
    If you sample at 10 Hz then you can identify frequencies in the range 0 - 5 Hz, which should be more than sufficient for (human) heart rate. – Paul R Jan 19 '16 at 08:31
  • Correct! i was cofusing units. I just realized that 1Bpm = 1/60 Hz. Thanks for your comment – Milorenus Lomaliza Jan 19 '16 at 08:51

0 Answers0