计算机工程与应用 ›› 2016, Vol. 52 ›› Issue (6): 167-171.

• 图形图像处理 • 上一篇    下一篇

基于ICA和EMD的生理信号提取

张言飞,欧阳健飞,姚丽峰   

  1. 天津大学 精密测试技术及仪器国家重点实验室,天津 300072
  • 出版日期:2016-03-15 发布日期:2016-03-17

Physiological signals extraction based on ICA and EMD

ZHANG Yanfei, OUYANG Jianfei, YAO Lifeng   

  1. State Key Lab of Precision Measuring Technology and Instruments, Tianjin University, Tianjin 300072, China
  • Online:2016-03-15 Published:2016-03-17

摘要: 提出一种新的获取人体生理参数的方法,用摄像头采集人脸的彩色视频,对人脸区域进行颜色通道分离和独立成分分析(ICA),获取有用信号。使用经验模态分解(EMD)的方法,把信号分解成可以反映出生命信息的固有模态函数(IMF),再根据所设计的提取准则,分别提取出较为准确的心跳和呼吸信号。用Bland-Altman法进行对比实验分析,结果表明,此方法具有一定的准确性和实用性。

关键词: 生理信号, 独立成分分析, 经验模态分解, 信号提取

Abstract: A new method for obtaining human physiological parameters is proposed. Useful signal is obtained from the color video of human face by color channels separation of face region and Independent Component Analysis(ICA). Then Empirical Mode Decomposition(EMD) method is used to decompose the signal into Intrinsic Mode Functions(IMFs) that can reflect the life information. According to the criterion of extraction, accurate heartbeat and respiration signals can be extracted from the IMFs. Bland-Altman method is applied to analyzing the experimental data. The results show that, this method is  of practicability and veracity.

Key words: physiological parameters, Independent Component Analysis(ICA), Empirical Mode Decomposition(EMD), signal extraction