使用网络摄像头估计生理参数的检测系统

A detection system that uses a webcam to estimate physiological parameters

目前基于IPPG技术的非接触式生理参数测量方法往往需要使用工业相机采集高帧率视频源以提高测量精度,但工业相机的使用门槛较高。针对此问题,文中设计了一款基于网络摄像头的非接触式生理参数检测算法,对网络摄像头提取到的IPPG信号进行特定的数字滤波和特征提取,能够测量人体心率和血氧饱和度。并且基于上述方法实现一款适用于PC端的生理参数检测系统。经过实验分析,该系统测量的心率误差为2.375bpm,血氧误差为0.405%,验证了消费者级摄像头采集人脸视频在测量人体生理参数方面的可行性。

At present,non-contact methods of measuring physiological parameters based on IPPG technology often require industrial cameras to capture high-frame-rate video sources for improving the accuracy of measurement,however,industrial cameras have a high threshold.To solve this problem,this paper designs a non-contact physiological parameters detection algorithm based on webcam,and specific filter and feature extraction are carried on IPPG signal extracted from the webcam that could measure the heart rate and oxygen saturation of human body.Based on the above methods,a system that detects physiological parameters suitable for PC system is realized.Through experimental analysis,the error of heart rate and oxygen saturation measured by this system are 2.375bpm and 0.405%respectively.The feasibility of using consumer-level camera to capture human face video for measuring human physiological parameters is demonstrated.