Non-contact respiration monitoring method based on visual calculation

Abstract

The invention discloses a non-contact respiration monitoring method based on visual calculation. The method comprises the following steps: acquiring a video image by using a camera; positioning a region of interest (ROI); graying processing is carried out; carrying out phase motion amplification processing on the respiratory motion by utilizing an innovatively improved phase motion amplification algorithm; extracting the respiratory signal after the phase motion amplification processing; kalman filtering processing is carried out on the extracted respiration signals; using an SVM classifier toclassify the respiratory signal waveform after Kalman filtering; judging whether the breathing is hypopnea breathing or not; in the case of non-low ventilation, a band-pass filter and fast Fourier transform are used for calculating the respiratory rate, the problems existing in a contact type detection method in the prior art are solved, the ultra-low-power-consumption non-contact type monitoringmethod based on visual calculation is provided, a conventional phase motion amplification method in the prior art is subverted, and the accuracy of phase motion amplification is improved. The accuracy of non-contact respiration monitoring results and respiration rate calculation is improved, and the method has high operability and practicability.

Description

technical field [0001] The invention relates to the technical field of medical respiration monitoring, in particular to a non-contact respiration monitoring method based on visual calculation. Background technique [0002] Breathing is a very important physiological activity of the human body. Sleep breathing signals such as the frequency, intensity, and rhythm of the human body can largely reflect the health status of the body and reveal the emergence of diseases. Therefore, monitoring of human sleep breathing has a very important medical value. significance. [0003] There are mainly two types of respiratory frequency detection methods: contact and non-contact. At present, medical respiratory monitoring is mainly based on contact monitoring. Complete the experiment under the condition of intervention, do not produce results because of psychological likes and dislikes and stimuli, if you use wearable or contact type, it may affect the experimental data, and contact type de...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to: in view of the problems existing in the contact detection method in the prior art and the technical problem that it is difficult to accurately monitor the breathing signal when the breathing is weak or the amplitude is small in the non-contact monitoring method, the present invention provides a method based on The non-contact breathing monitoring method of visual calculation, the technical solution provided by the present invention not only solves the problems existing in the contact detection method, but also aims at the non-contact breathing monitoring method in the prior art that is caused by weak breathing signals or small amplitude. For the problem of inaccurate monitoring results of respiratory signals, an innovative solution is provided, which innovatively improves the phase motion amplification method in the prior art, and applies this method to the non-contact respiratory monitoring process for the first time, to monitor respiratory motion Perform phase motion amplification processing to realize real-time monitoring of the non-contact respiratory monitoring method, and overcome the problem of inaccurate respiratory monitoring results caused by the non-contact monitoring method in the prior art due to weak breathing or small amplitude

Images