Non-contact type heart rate measurement method based on CMOR wavelets

Abstract

The invention discloses a non-contact type heart rate measurement method based on CMOR wavelets. The method comprises steps as follows: firstly, a 60s testee face video in an AVI format is recorded with natural light as a light source, the recorded video is detected, the detected human face area is outlined and cut out by a polygon and set as a reconstructed face image, and then the eyes, the nose, the mouth and the center point of the face are selected as tracking characteristic points; the reconstructed face image is converted into YCbCr space for skin detection; Cg channel green concentration deviation is sequentially recorded for other frames except the face in the recorded video; the Cg channel green concentration deviation is taken as an original PPG signal and subjected to band-passfiltering processing, noise signals outside pass bands are filtered out, and a target heart rate signal is obtained; and the target heart rate signal is subjected to CMOR wavelet analysis to obtain awavelet time-frequency graph. The heart rate can be accurately measured in real time, and the defects that a conventional peak interval algorithm has larger error and a heart rate parameter as a dominant component in a period of time can only be extracted by Fourier transform are overcome.

Description

technical field [0001] The invention relates to a non-contact heart rate measurement technology, in particular to a non-contact heart rate measurement method based on CMOR wavelet. Background technique [0002] Along with the continuous development of society, people's living standard improves day by day, therefore, health problem more and more causes people's attention. Heart rate is an important parameter to measure the health index of human body, which can reflect the health level of human body. In recent years, cardiovascular disease, high blood pressure, coronary heart disease, diabetes, etc. have come one after another, not only causing great harm to people's physical health, but also causing great distress to people's psychology. Therefore, disease prevention and Control is very important to our life. Regular heart rate measurement can effectively prevent the occurrence of various cardiovascular diseases and has an irreplaceable role in human health. [0003] The cu...

AI Technical Summary

Problems solved by technology

However, the existing non-contact heart rate measurement technology has the problem of large measurement error. The reasons are: first, the signal strength extracted through the face color change is very weak, the signal-to-noise ratio is low, and it is easily disturbed by the external environment (For example, if the ambient light changes, the head movement is too large to accurately track the face, etc.), the quality of the target heart rate signal obtained by using the blind source analysis algorithm is poor, and it cannot be determined where the target heart rate signal is located in the signal after blind source separation. One part; the second is to perform Fourier transform on the extracted heart rate signal, and only the heart rate parameters that account for a period of time can be observed, which cannot meet the real-time detection requirements; the third is to obtain the heart rate parameters by calculating the peak interval algorithm, but Due to the existence of noise interference, there will be peak shift, missed detection, multiple detection, etc., which will seriously affect the accuracy of heart rate parameter measurement

Images