Research On Pulse Wave Extraction Method Based On Video Amplification

Cardiovascular disease is a common disease that endangers human health.It is important to detect cardiovascular health status through non-clinical means.Pulse wave is one of the important indexes to reflect the cardiovascular health status of human body.At present,video non-contact pulse wave extraction research can be divided into two categories: one is based on the change of facial cheek color pulse wave extraction,the other is based on the wrist radial artery small pulse changes pulse wave extraction.The changes of radial artery movement can reflect the real situation of cardiovascular system more than the changes of facial color.Therefore,the extraction of radial artery pulse wave has higher research value.According to the physiological characteristics of the radial artery at the wrist,this paper proposes a pulse wave extraction algorithm based on video motion amplification,which is innovatively summarized as follows:(1)In view of the existing video method does not consider the problem of noise signal suppression in the process of pulse wave signal acquisition,this paper proposes a radial pulse wave extraction algorithm based on Eulerian video amplification signal selection.Based on the existing algorithms,Laplace pyramid strategy is used to extract more complete multi-scale interested signals.According to the size of signal-to-noise ratio,the region with high signal-to-noise ratio is selected as the region of interest to avoid the interference of noise signal,so as to finally realize the heart rate measurement of high reliable pulse signal.(2)A pulse wave extraction system based on video amplification is designed and implemented.The system takes the radial artery as the extraction object,carries out video data collection on the radial artery of the wrist through the ordinary camera,and then measures pulse and heart rate through the algorithm proposed in this paper.The development of the system USES the software engineering development specification,and the system test shows that it has better robustness and accuracy.In order to verify the accuracy and robustness of the non-contact pulse wave detection algorithm proposed in this paper,video of radial artery in the wrist of human body is collected for experiment.Experimental results show that the proposed method is more accurate and robust than the existing methods.In addition,the system implemented in this paper has a lower time complexity.