Method for eliminating influence of body motion in vital sign detection based on optical flow method and LMS algorithm

Abstract

The invention discloses a method for eliminating influence of body motion in vital sign detection based on an optical flow method and an LMS algorithm, wherein the method comprises the following steps: obtaining a vital sign signal from an echo signal collected by a radar sensor, extracting a breathing signal and a heartbeat signal from the signal; collecting a video image of an object to be tested by a video sensor, and selecting an ROI region; obtaining an optical flow vector of each pixel in the ROI region of each frame of image by the optical flow method, and calculating the position change of the object to be measured relative to the radar sensor according to the obtained optical flow vectors; and according to the position change, eliminating the body motion of the object to be testedby the LMS algorithm, so as to obtain a better vital sign signal. The method is realized by combining the single radar sensor with the video sensor, and has low cost; at the same time, the influencecaused by random movement of a human body can be effectively eliminated by combining the optical flow method with the LMS algorithm, an additional testing device is not needed to be additionally addedon the object to be tested and the method is effective, feasible and reliable in performance.

Description

technical field [0001] The invention belongs to the field of vital sign monitoring, in particular to a method for eliminating the influence of body movement in vital sign detection based on an optical flow method and an LMS algorithm. Background technique [0002] Since the 1970s, the research on Doppler radar vital signs detection has attracted more and more attention. People can use Doppler radar to detect the respiration and heartbeat of the subject in a non-contact manner. Although many studies have been done to develop this technique and many difficulties have been overcome in the past few decades, random body motion is still an intractable problem in Doppler radar detection. Usually, in the process of vital signs detection, people’s random body movement frequency will cover the frequency of breathing and heartbeat. In addition, the amplitude of body movement is often much larger than the chest undulation caused by breathing and heartbeat. unattainable. Due to the exi...

AI Technical Summary

Problems solved by technology

Although many studies have been done to develop this technique and many difficulties have been overcome in the past few decades, random body motion remains an intractable problem in Doppler radar detection

Usually, in the process of vital signs detection, people’s random body movement frequency will cover the frequency of breathing and heartbeat. In addition, the amplitude of body movement is often much larger than the chest undulation caused by breathing and heartbeat. Unattainable

Due to the existence of unpredictability, in real life, random body motion will have an inevitable impact on the detection of Doppler radar vital signs

[0003] Existing body motion removal methods mainly include using the Empirical Mode Decomposition (EMD) method to remove the subject’s body motion interference, but this method is more suitable for motion pattern classification. The effect is not good; the multi-radar sensor array requires the subject to move the body between the two radar sensors. This method relies on the baseband signal, and the use of multiple radar sensors at the same time will greatly increase the complexity and cost of the system; in addition, There is also a system based on an injection-locked radar array to eliminate random body movements, but this system requires scanning the carrier frequency of the radar or adjusting the target position, and the overall process is very complicated

The cost and complexity of the system restrict the development of the above methods, and it is a major challenge in this field to achieve high accuracy and low complexity and cost.

Images