Research On Heart Rate Measurement Based On Face Image Enhancement

With the improvement of health care awareness,people pay more and more attention to the daily detection of heart rate.Heart rate measurement based on face image has become a research hotspot because of its advantages of no contact,convenient measurement and simple equipment.Due to the influence of noise and unconsciousness shaking,the existing methods have some problems,such as low accuracy and adaptability,so heart rate measurement based on face image becomes a challenging task.In order to improve the accuracy of heart rate measurement,this paper studies the optimization of each link of heart rate measurement based on face image,and proposes a new method of heart rate measurement based on face image.The main work is as follows:1)Aiming at the problem that the signal amplitude of face video centrality is weak and often covered by noise,Euler image color amplification algorithm is used to amplify the 0.85Hz～2Hz frequency band where the heart rate is,so as to highlight the heart rate signal and improve the signal-to-noise ratio.2)Aiming at the problem that the frequency band of heart rate signal contains a lot of noise,the VMD algorithm is applied to the filtering process of time series.According to the difference of center frequency,the original signal is decomposed into the sum of several discrete sub signals with specific sparsity.Each sub signal is called a mode.After the separation,the mode whose center frequency does not conform to the normal heart rate range is eliminated first,and then the mode which is more likely to be the heart rate signal is highlighted by multiplying the coefficient,and the mode with the maximum power after multiplying the coefficient is retained.After that,the comparative experiments of decomposition effect are carried out under different parameter settings,and the reasonable parameter values are given.3)Then,the X_sminαY_s algorithm is used to synthesize the heart rate signal sequence from the three channels to reduce the interference of motion artifacts and improve the measurement robustness.4)Using DEAP database and another database of my own,a complete heart rate measurement comparative experiment was carried out according to the method in this paper.The experimental results show that the method proposed in this paper achieves better measurement accuracy and robustness than the traditional algorithm,and achieves good results.