Abstract
Non-contact physiological parameters detection has become an important but challenging task. This paper proposes a novel framework for Non-contact physiological parameters detection, which is based on MTCNN (Multi-task convolutional neural network) and EVM (Eulerian Video Magnification). The MTCNN is applied to compensate negative effects of the pseudo motion for improving the efficiency of face detection. The Laplacian pyramid of EVM is added to our framework to perform motion amplification to achieve simultaneous extraction of heart rate and respiration rate, which improves the accuracy of non-contact physiological parameter detection. Then we use FFT to analyze the relevant frequency bands in time domain and calculate the heart rate and breathing rate. In order to improve efficiency, GPU acceleration is implemented under the TensorRT framework. The experimental results based on data sets DUSS (detection under slight shaking) and DUO (detection under occlusion) show the effectiveness of this method. The average error of the heart rate and respiration rate are −0.1 and 0.6, respectively. Our method achieves comparable performance to finger clip oximeter with a good consistency, even in the case of wearing a mask.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Wang, J., Xue, H., Lu, H., et al.: Non-contact physiological signal detection technology. China Med. Equip. 11, 5–8 (2013)
Liu, Y.: Non-contact heart rate and respiratory rate detection based on FastICA algorithm. Nanchang University, Nanchang (2015)
Liu, H., Wang, X., Chen, G.: Non-contact fetal heart rate detection. Comput. Syst. Appl. 28(8), 204–209 (2019)
Su, P., Liang, X., Liang, Y.: Non-contact heart rate measurement method based on Euler image magnification. J. Comput. Appl. 35(3), 916–922 (2018)
Li, J.: Research on pulse wave extraction method based on video amplification. Hefei University of Technology, Hefei (2019)
Ma, L.: Research on non-contact physiological parameter detection technology based on ordinary camera. Shandong University, Shandong (2017)
Deng, Q.: Research and implementation of heart rate detection based on face video images. Wuhan University of Technology, Wuhan (2017)
Hui, P.: Research on non-contact heart rate measurement method based on face image. Southeast University, Nanjing (2017)
Viola, P., Jones, M.: Rapid object detection using a boosted cascade of simple features. In: Proceedings of CVPR, vol. 1, pp. 511 (2001)
Liu, X., Cui, H., Liu, Z., et al.: Human body sign detection based on Doppler radar sensor. Sens. Microsyst. 2020(39), 137–139 (2020)
Cui, C., Zhao, Q.: Design of vital signs detection system based on micro-Doppler Radar. Control Eng. 26, 105–107 (2019)
Zhou, Y., Pang, L., Zhang, H., et al.: Experimental research on non-contact life detection based on bio-radar technology. Sci. Technol. Vis. 181–182
Zhang, Y., Lu, H., Liang, F., et al.: Current status and application prospects of non-contact battlefield injury detection technology. Med. Med. Equip. 40(07), 99–103 (2019)
Cai, S., Men, X., Wang, S., et al.: Research on rapid and accurate measurement of temperature in dense populations during the epidemic of new coronavirus pneumonia. West China Med. 35, 385–390 (2020)
Ge, Z.: Research on medical infrared thermometer and its key technology. Changchun University of Science and Technology, Changchun (2019)
Ku, H., Dong, W.: Face recognition based on MTCNN and convolutional neural network. Front. Sig. Process. 4(1), 37–42 (2020)
Wu, J., Chen, S.: An improved MTCNN face detection algorithm. Softw. Guide 18(12), 78–81 (2019)
Cai, K.: Research on non-contact heart rate measurement technology. Nanjing University of Posts and Telecommunications (2019)
Liu, Y.: Human heart rate measurement based on facial video. Shanghai Normal University, Shanghai (2018)
Yi, S., Zhu, J., Jing, H.: Research on face occlusion recognition based on MTCNN in railway face brushing scene. Comput. Simul. 37(5), 96–99 (2020)
Chen, D.: Research on non-contact heart rate measurement based on video. South China University of Technology, Guangdong (2016)
Acknowledgments
This work was supported by the Project (19YFZCSF01150) in the Science & Technology Pillar Program of Tianjin, in part by the Tianjin Natural Science Foundation of China under Grant 18JCZDJC40300, in part by Innovation cultivation Foundation 19-163-12-ZT-006-007-06.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Zhao, J., Zhang, W., Chai, R., Wu, H., Chen, W. (2021). Non-contact Physiological Parameters Detection Based on MTCNN and EVM. In: Sun, F., Liu, H., Fang, B. (eds) Cognitive Systems and Signal Processing. ICCSIP 2020. Communications in Computer and Information Science, vol 1397. Springer, Singapore. https://doi.org/10.1007/978-981-16-2336-3_48
Download citation
DOI: https://doi.org/10.1007/978-981-16-2336-3_48
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-2335-6
Online ISBN: 978-981-16-2336-3
eBook Packages: Computer ScienceComputer Science (R0)