Research on Non-contact Heart Rate Signal Detection Based on Video Image

In this paper, the heart rate signal is extracted from the video containing the frontal face using the principle of Cosmography. The separation of the face and the video background area is used to remove the motion artifacts generated by the motion of the subject and the interference of the physiological motion. Furthermore, removing the motion artifacts for the ROI region and extracting the original signals of the ROI region and the background region respectively, and combining the normalized adaptive filtering algorithm to remove the interference of the ambient light to the heart rate signal, then removing the heart rate signal by the time domain filtering. Heart rate values were extracted using periodic analysis. It is found through experiments that the heart rate test is more efficient under the conditions of ambient light changes, indicating that the scheme has better stability under the environmental illumination changes.