Research On Non-contact Heart Rate Measurement Based On Video Sequences

With the accelerated pace of life,the incidence rate of cardiovascular diseases has considerably increased,becoming one of the major factors threatening people’s health.As a feedback indicator of physical function,heart rate plays crucial role in the prevention and diagnosis of cardiovascular diseases.For a long time,heart rate was measured by means of traditional contact methods.However,the contact method will bring disadvantages such as limited movement and skin compression during continuous monitoring,so it is not practical enough in dynamic heart rate monitoring and the nursing of special personnel.In recent years,the Imaging Photoplethysmography(IPPG)has provided a contactless type of heart rate measurement,which continues to gain attention due to convenient measurement and wide application.In this paper,we proposed a method for measuring heart rate based on the LAB color space to solve the interference in IPPG measurements,and the main work is as follows:Based on the VJ face detection operator to accurately locate faces in real time,and the elliptical skin model is used to obtain the skin regions.Then,the conversion from RGB color space to LAB color space is performed to separate the luminance signal,and the smoothness prior approach is employed to remove the stationary artifacts in the raw signals of A channel and B channel.On this basis,the difference between A channel and B channel is introduced as pseudo-pulse signal.Next,the ensemble empirical mode decomposition algorithm is used to decompose the pseudo-pulse wave to obtain a set of Intrinsic Mode Functions(IMF).The IMF with the highest energy ratio and peak ratio is selected,and the heart rate(HR)is estimated based on the frequency corresponding to the peak.Finally,we conducted experimental analysis through 90 videos taken by 30 volunteers.Experimental results show that the mean absolute deviation of heart rate measured based on the method in this paper is only 2.54 beats/min(bpm),with 95% confidence interval of-7.14 bpm to 3.40 bpm.