The Noninvasive Blood Pressure Measurement Based on Facial Images Processing

Abstract

This paper proposes a noninvasive method of physiological parameter measurement with the collected facial video images for continuous online monitoring. Blood pressure (BP) is one of the important physiological indexes for human health prognoses. Based on the photoplethysmograph (PPG) technology, the variation of the facial skin color along with the heart beating can thus be captured via the camera. So the related BP information can be extracted from the blind source signal separation. The detailed procedure of the BP measurement from the facial video is explained, including video collection, face tracking, signal pre-processing, and BP signal extraction. The systolic and diastolic BP model can be derived based on the radial resonance theory and the involved parameters are fitted via the sampling data. Experiments are performed in different scenarios to discuss the impact factors on the BP detection together with error compensation analysis. The experimental results demonstrate that the proposed BP online measurement can be used as an indicator to monitor the status of the human health continuously.