| Cardiovascular disease is the most deadly disease.Measuring heart rate and blood pressure is of great significance for the prevention and treatment of cardiovascular disease.The light reflected by the human skin changes with the heartbeat.The camera can capture this faint change and further obtain information such as heart rate,blood pressure,and blood oxygen.This technology is called Imaging Photo Plethysmo Graphy(IPPG).In recent years,the non-contact monitoring system based on IPPG technology has become a research hotspot.The core of the non-contact monitoring system is the non-contact physiological parameter measurement algorithm.Compared with the traditional contact monitoring system,the non-contact monitoring system has the advantages of no sense of restraint,wide application scenarios,and low cost.The change in the skin's reflected light caused by the heartbeat is very weak,while the changes of human body movement and ambient light will produce very strong noises.How to effectively eliminate the noise caused by human body movements and ambient light changes is an unsolved problem at present.In this paper,aiming at the shortcomings of existing methods with poor denoising effect,we studied how to obtain a higher signal-to-noise ratio IPPG signal and improve the accuracy of heart rate and blood pressure measurement.Based on the non-contact physiological parameter measurement algorithm,this paper designs and implements the mobile terminal non-contact monitoring system.The work of this paper mainly includes the following aspects:1.This paper analyzes the noise caused by the rigid and non-rigid motion of the face in the IPPG signal,and proposes a region of interest extraction algorithm based on feature tracking.This algorithm combines facial keypoint detection and feature tracking to capture the cheek area as the region of interest.This algorithm can effectively reduce the noise generated by the rigid and non-rigid motions of the face,and improve the SNR of the IPPG signal.2.In order to further eliminate the noise caused by the change of ambient light,this paper proposes a heart rate measurement algorithm based on wavelet packet denoising.The algorithm converts the region of interest into YIQ color space,calculates the mean value of Q channel,and obtains the original IPPG signal,which reduces the noise caused by the change of ambient brightness.Wavelet packet decomposition and threshold denoising are used to further remove the noise.An adaptive crest detection algorithm was designed to detect the number of effective peaks in the IPPG signal to obtain the heart rate.Experiment results show that,compared with the two existing methods,the non-contact heart rate measurement algorithm proposed in this paper can effectively eliminate the noise caused by human movement and environmental light changes,and the measurement error is smaller and the robustness is stronger.3.This paper explores the application of IPPG technology in blood pressure measurement.This article gathered 30 volunteers.First,the non-contact method was used to measure the pulse waves of the head and hands of the volunteers,and a data set containing 150 data was established.Then,the blood pressure was measured by two methods,linear regression and neural network.The Bland-Altman method is used to analyze the performance of the algorithm.Experiment results show that the measurement results of the two blood pressure measurement methods are in good agreement with the electronic sphygmomanometer,and the measurement results of the neural network method are better than the linear regression method.4.A non-contact monitoring system is designed in this paper.The system consists of a mobile terminal,a server terminal,and an embedded terminal,and the mobile terminal part is developed.This system has a wide range of application scenarios,which can realize non-contact heart rate measurement and monitoring,remote video,smart doctor and other functions. |
PDF Full Text Request