Research On Techniques Of The Noncontact Detection Of Physiological Parameters Based On Ordinary Camera

With the economic development and the rapid urbanization process,the style of working and the lifestyles of citizens have big changes,the incidence of cardiovascular disease has increased significantly coupled with the aging tendency.Heart rate and oxygen saturation are physiological parameters related to cardiovascular disease closely,and they play important roles in clinical diagnosis and daily testing.However,direct contact with the respondents is required when using the current widely used detection of physiological parameters,which is not suitable for newborns or buns and scalds;Furthermore,some equipment is too complex to be used for daily testing.Imaging Photoplethysmography(IPPG)is a noncontact physiological parameter detection technique developed in recent years.IPPG technology has the advantages of non-contact,easy operation and loose requirements of hardware,which attracts wide concern of researchers in the medical and instrumentation measurement field.How to make the equipment simpler and how to get more accurate and comprehensive physiological parameters are the main research directions.This dissertation introduces the existing methods of physiological parameters'detection,discusses the shortcomings of contact detection,and analyzes the optics of human skin and the principle of IPPG.Furthermore,it summarizes the methods of measuring the physiological parameters using IPPG,and elaborates the characteristics of blood volume pulse.Then,based on IPPG,combined with Eulerian video magnification and wavelet transform,using ordinary cameras and laptops,we measure heart rate and oxygen saturation without direct contact with skin under ambient light.The work of the dissertation is as follows:1.Design and realize the project of detection of heart rate based on ordinary camera,using Eulerian video magnification and wavelet transform.In this part we demonstrate the feasibility of using the ordinary CCD camera and lens to realize the heart rate,compare the difference from different regions of interest and choose the rectangular area of the forehead as the region of interest,then use the Eulerian video magnification to magnify varies in brightness of face and use wavelet transform to remove noise,get the blood volume pulse waveform,use FFT to get heart rate at last.The experimental results show that the project has high accuracy and can be suitable for daily testing.2.Design and realize the project of detection of oxygen saturation based on a dual camera with filter,measure oxygen saturation without direct contact with skin under ambient light,do not need specific light source.In this dissertation,we analyze the principle of dual wavelength method for detecting oxygen saturation,and choose filter's wavelength.Then use a dual camera with 660nm and 525nm filter to recorded videos,use the Eulerian video magnification and wavelet transform to get the waveform of volume pulse under different wavelengths,then average the peak-valley values of different waveform,calculate the value of oxygen saturation.The experimental results show that there are systematic errors compared with the reference values.The dissertation demonstrates the cause of the system errors and makes a rough correction of the results.The corrected results are in good agreement with the reference values.But the precise correction needs to re-determine the empirical parameters,and we can't have a try in the laboratory environment.