基于视频的人体心率检测算法研究

Research on Human Heart Rate Detection Algorithm Based on Video

传统的心率检测方法有侵入式和非侵入接触式两种测量方法,但是会造成不便。论文提出一种基于IPPG(Image photoplethysmography)技术的非接触式人体心率测量方法,从人体的裸露皮肤提取脉搏波信号,在脉搏波信号中提取人体的心率值。通过摄像头采集人体面部的信息,提取感兴趣区域的灰度均值后,经处理得到人体的心率。论文还通过实验调整了心率检测系统中的感兴趣区域、采样点数和图像颜色格式,使检测精度得到提升。利用论文方法测量人体心率,与指夹式血氧仪的测量数据进行比对,结果的平均绝对误差为1.91次/min和均方根误差为2.15次/min。

Traditional heart rate detection methods include invasive and non-invasive contact measurement methods,but it will cause inconvenience.In this paper,a non-contact human heart rate measurement method based on IPPG(image photometry)technology is proposed.The pulse wave signal is extracted from the human exposed skin,and the human heart rate is extracted from the pulse wave signal.The human face information is collected by the camera,and the gray mean of the region of interest is extracted,and the human heart rate is obtained after processing.This paper also adjusts the region of interest,sampling points and image color format in the heart rate detection system through experiments to improve the detection accuracy.Compared with the measured data of finger clip oximeter,the average absolute error and root mean square error of the results are 1.91 beats/min and 2.15 beats/min.