基于自适应变分模态分解的视频心率检测算法

Video heart rate detection algorithm based on adaptive variational mode decomposition

视频心率检测算法不够鲁棒,光照和运动等因素严重影响心率信号的提取。针对此问题,提出一种自适应变分模态分解算法。利用经验模态分解(Empirical mode decomposition,EMD)对远程光学描记术信号(remote photoplethysmography,rPPG)进行分解获取信号的模态数,设计自适应变分模态分解(Variational mode decomposition,VMD)算法,分解rPPG信号获取不同的频率和带宽的模态,根据心率的频率范围和模态的中心频率合成心率信号,然后对心率信号进行功率谱分析计算心率,有效解决了VMD采用固定模态数分解的不足。在公开数据集PURE和UBFC上的实验结果表明,所提方法心率检测平均绝对误差小于5;与经验模态分解等方法相比,均方根误差减小了2.19。

The video heart rate detection algorithm is not robust enough,illumination and motion factors seriously affect the heart rate signal extraction.Aiming at this problem,an adaptive variational mode decomposition algorithm is proposed.Specifically,empirical mode decomposition(EMD)is used to decompose the remote photoplethysmography(rPPG)signal to obtain the modal number of the signal.The adaptive variational mode decomposition(VMD)algorithm was designed to decompose the rPPG signal to obtain the modes with different frequencies and bandwidths.According to the frequency range of the heart rate and the modal center frequency,the heart rate signal is synthesized,and then the heart rate is calculated by power spectrum analysis of the heart rate signal,which effectively solves the shortage of the VMD by using fixed mode number decomposition.The experimental results on the public datasets PURE and UBFC show that the mean absolute error in heart rate detection of the proposed method is less than 5.Compared with empirical mode decomposition,the root mean square error is reduced by 2.19.