计算机工程与应用 ›› 2022, Vol. 58 ›› Issue (9): 271-278.DOI: 10.3778/j.issn.1002-8331.2011-0284

• 工程与应用 • 上一篇    下一篇

现实场景中非接触式心率检测方法研究

牟睿,陈鲸,杨学志,吴克伟,方帅   

  1. 1.合肥工业大学 计算机与信息学院,合肥 230009
    2.工业安全与应急技术安徽省重点实验室,合肥 230009
    3.合肥工业大学 科研院,合肥 230009
  • 出版日期:2022-05-01 发布日期:2022-05-01

Research on Non-Contact Heart Rate Detection in Real Scene

MOU Rui, CHEN Jing, YANG Xuezhi, WU Kewei, FANG Shuai   

  1. 1.School of Computer and Information, Hefei University of Technology, Hefei 230009, China
    2.Anhui Province Key Laboratory of Industry Safety and Emergency Technology, Hefei 230009, China
    3.Scientific Research Institute of Hefei University of Technology, Hefei 230009, China
  • Online:2022-05-01 Published:2022-05-01

摘要: 现有的非接触式心冲击描记术(ballistocardiography,BCG)利用运动追踪捕捉头部微弱运动,存在着光照干扰、运动干扰等问题。针对该问题,提出了一种基于方向可调滤波器的头部微小振动检测方法(steerable filters schematic detection BCG,SD-BCG),以实现非接触式心率检测。对输入视频进行人脸检测,并框选出目标检测区域;设计一种基于方向可调滤波器的振动信号提取方法,提取目标区域的相位差信号,并针对相位差信号进行带通滤波和主成分分析,滤除干扰噪声并建立BCG信号;通过傅里叶变化将BCG信号转化至频域,从频谱密度函数中提取心率值。实验结果表明该方法能够适应不同光照和不同运动干扰情况,其心率检测准确率优于现有的运动追踪捕捉BCG信号提取方法。

关键词: 非接触式, 心冲击描记术, 头部微小振动, 方向可调滤波器

Abstract: Existing non-contact method to detection ballistocardiography(BCG) uses motion tracking to capture faint head movement, there are light interference, movement interference and other issues. To solve this problem, a faint head movement detection method based on steerable filters schematic(SD-BCG) is proposed to realize non-contact heart rate detection. Face detection is performed on the input video, and the target detection area is selected. A vibration signal extraction method based on directional adjustable filter is designed to extract the phase difference signal in the target area, and the phase difference signal is then bandpass filtered and principal component analyzed to filter out the interference noise and establish BCG signal. The BCG signal is transformed into the frequency domain by Fourier change, and the heart rate is extracted from the spectral density function. The experimental results show that this method can adapt to different conditions of light and motion interference, and its heart rate detection accuracy is better than the existing motion tracking capture BCG signal extraction method.

Key words: non-contact, ballistocardiography(BCG), microvibration of the head, steerable filters schematic