中文版 | English
CSpace
Non-contact heart rate measurement based on facial video
Zhou, Nan1,2; Huang, Lei1; Huang, Jifeng2; Pan, Tao3; Ni, Yepeng4; Shan, Lianhai5
2018
摘要Heart rate measurement is an important means of monitoring the physiological status of the human body. A non-contact heart rate measurement method is proposed by analyzing the change of pixels. Multi-filter and sub-region method are adopted to solve the problem of the complex heart rate signal extraction result. As the independence of the RGB color model, the green component signals can be taken from the mixed signals by using Fast ICA. The main peak noise is removed by the wavelet filtering and the band-pass filter, and the heart rate is obtained in the analysis of the energy spectrum by using the Fourier transform. The facial organs influence on the extraction of the heart rate signal is effectively reduced by dividing the face region, and the part of the face to best reflect the heart rate is further determined. The experimental results show that this method can be convenient and accurate to achieve the heart rate measurement, and the proposed division method of the most appropriate facial part to extract the heart rate enhances the application of the heart rate measurement. © 2018 The authors and IOS Press. All rights reserved.
语种英语
DOI10.3233/978-1-61499-927-0-818
会议(录)名称4th International Conference on Fuzzy Systems and Data Mining, FSDM 2018
页码818-823
收录类别EI
会议地点Bangkok, Thailand
会议日期November 15, 2018 - November 16, 2018