计算机工程与应用 ›› 2022, Vol. 58 ›› Issue (6): 58-68.DOI: 10.3778/j.issn.1002-8331.2110-0050

• 热点与综述 • 上一篇    下一篇

基于视频的生理参数测量技术及研究进展

陈嘉涛,张泓凯,黄燕平,蓝公仆,许景江,秦嘉,安林   

  1. 1.佛山科学技术学院 机电工程与自动化学院,广东 佛山 528000
    2.佛山科学技术学院 物理与光电工程学院,广东 佛山 528000
    3.广东省“珠江人才计划”引进创新创业团队 广东唯仁医疗科技有限公司,广东 佛山 528051
  • 出版日期:2022-03-15 发布日期:2022-03-15

Video-Based Physiological Parameters Measurement Technology and Research Advances

CHEN Jiatao, ZHANG Hongkai, HUANG Yanping, LAN Gongpu, XU Jingjiang, QIN Jia, AN Lin   

  1. 1.School of Mechatronic Engineering and Automation, Foshan University, Foshan, Guangdong 528000, China
    2.School of Physics and Optoelectronic Engineering, Foshan University, Foshan, Guangdong 528000, China
    3.Innovation and Entrepreneurship Teams Project of Guangdong Pearl River Talents Program, Guangdong Weiren Meditech Co., Ltd., Foshan, Guangdong 528051, China
  • Online:2022-03-15 Published:2022-03-15

摘要: 生理体征参数的临床监测对于个体的健康管理、疾病的预防与跟踪具有重要的意义。远程光电容积描计法(remote photoplethysmography,rPPG)是一项新兴的、基于摄像头的、无接触方便快捷的生理体征参数测量技术。血管内部血液的流动造成人体表皮颜色的微弱变化,通过计算机对这一周期性变化的信息进行放大提取,便可以分析出与心脏运作相关的体征参数,如心率、血压等。RPPG自2008年提出之后得到了飞速的发展,其准确率、鲁棒性、实用性都得到了大幅度的提高。RPPG的检测流程主要是从图像传感器获取人体面部数据开始,随后基于图像及信号处理技术完成微弱脉冲波形的提取及相关体征参数的计算。从rPPG的基本检测流程出发,对rPPG的硬件采集设备进行介绍,重点描述从rPPG中提取心率的算法,进一步介绍rPPG相关的应用研究,就未来rPPG可能的研究方向进行了展望。

关键词: 远程光电容积描计法, 无接触测量, 生理体征参数, 光照, 运动, 深度学习

Abstract: The clinical monitoring of physiological parameters is of great significance for individual health management, disease prevention and tracking. Remote photoplethysmography(rPPG) is an emerging, camera-based, non-contact, fast and convenient technology for measurement of physiological parameters. The blood flow inside the microvascular vessels causes the slight changes in the color of the human skins. By amplifying and extracting this periodic change information, the physical parameters related to the function of the heart such as the heart rate and blood pressure can be analyzed. RPPG technology has developed rapidly since it was proposed in 2008, and its accuracy, robustness, and practicability have been greatly improved up to date. The rPPG detection process mainly starts with the acquisition of human facial data from the image sensor, and then completes the extraction of pulse waveforms and the calculation of related physical parameters based on image and signal processing technology. Starting from the basic detection process of rPPG, this article firstly introduces related hardware setup and processing algorithms, then presents its recent applications and finally the possible research directions of rPPG in the future are discussed.

Key words: remote photoplethysmography(rPPG), non-contact measurement, physiological parameters, illumination, motion, deep learning