KLT Algorithm for Non-contact Heart Rate Detection Based on Image Photoplethysmography

Heart rate is one of the four vital signs that has the most obvious response to changes in human health status. This paper applied a non-contact heart rate detection method based on image Photoplethysmography (IPPG). Firstly, RGB color images of channels were collected by ordinary general cameras, and the region of interest (ROI) was located by face recognition. Then KLT algorithm is used to calculate the target offset of the region of interest, and the ROI is modified. After that, RGB channel signals are converted to Lab color space to extract Blood Volume Pulse (BVP) signals. Finally, complex Morlet 4-2 (CMOR4-2) was used for time-frequency analysis after BVP signal filtering, and heart rate was calculated according to time-frequency analysis. Experimental results show that the accuracy of this method is 96.5% in a relatively resting environment. The system can be used in a wide range of scenarios and can detect heart rate in a general home environment. This method has the characteristics of non-trauma, no need to wear the relevant sensor, easy to use, and has a broad prospect of popularization in daily application.