CN-115713705-B - Pulse wave signal extraction method, pulse wave signal extraction device, storage medium and electronic equipment

CN115713705BCN 115713705 BCN115713705 BCN 115713705BCN-115713705-B

Abstract

The invention discloses a pulse wave signal extraction method, an extraction device, a storage medium and electronic equipment, wherein the pulse wave signal extraction method comprises the steps of collecting video data of a detected target, wherein the video data comprises a plurality of frames of video images and time stamps corresponding to the video images; the method comprises the steps of determining an information value of each frame of video image, obtaining vibration signals based on time sequences according to the information value and corresponding time stamps, inputting the vibration signals based on the time sequences into a preset space vector decomposition model, outputting a pulse wave signal space matrix, and carrying out normalization processing on the pulse wave signal space matrix to obtain pulse wave signals of a tested target. Therefore, the pulse wave signal extraction method can obtain the pulse wave signal of the pure measured object which is comparable to the contact pulse wave sensor.

Inventors

TAN CHUNJIE
WENG WENJUAN
YANG CHENGYUE

Assignees

安徽五域安全技术有限公司

Dates

Publication Date: 20260505
Application Date: 20210820

Claims (9)

1. The pulse wave signal extraction method is characterized by comprising the following steps: collecting video data of a detected target, wherein the video data comprises a plurality of frames of video images and time stamps corresponding to the video images; Determining an information value of each frame of video image, and obtaining a vibration signal based on a time sequence according to the information value and a corresponding time stamp; before inputting the vibration signal based on the time series into the preset space vector decomposition model, the method further comprises: Performing frame inserting processing on the vibration signal based on the time sequence according to the time stamp; The frame inserting processing is carried out on the vibration signal based on the time sequence according to the time stamp, and the frame inserting processing comprises the following steps: calculating the difference value of time stamps of any two adjacent information values in the vibration signal based on the time sequence; if the difference value is larger than a first preset time, determining that the vibration signal based on the time sequence needs to be subjected to frame inserting processing, and determining the number of frames needing to be inserted according to the difference value; Establishing a primary fitting function according to signal values at the left side and the right side of the frame inserting position and corresponding time points thereof, determining a middle time point to be inserted according to the missing time length of the time points at the left side and the right side, substituting the middle time point into the primary fitting function to obtain a corresponding signal value to be inserted, and inserting the signal value to be inserted and the corresponding time stamp thereof into the vibration signal based on the time sequence; inputting the vibration signals based on the time sequence into a preset space vector decomposition model, and outputting a pulse wave signal space matrix; And carrying out normalization processing on the pulse wave signal space matrix to obtain a pulse wave signal of the tested target.
2. The method for extracting a pulse wave signal according to claim 1, wherein determining the information value of the video image includes: Preprocessing the video image to obtain an interested region of the video image; performing RGB color vector decomposition on the region of interest to obtain an RGB color matrix; and calculating an average value of each color component in the RGB color matrix to obtain a corresponding information value.
3. The method for extracting pulse wave signals according to claim 1, wherein said obtaining a time-series based vibration signal from the information values and the corresponding time stamps comprises: Combining the information values of the video images according to time sequences to obtain a vibration signal matrix, and combining the time stamps of the video images according to time sequences to obtain a time stamp matrix; and combining the vibration signal matrix with the time stamp matrix to obtain the vibration signal based on the time sequence.
4. The method for extracting pulse wave signals according to claim 3, wherein, After the vibration signal matrix is obtained, the method further comprises inputting the vibration signal matrix into a preset first filter to perform filtering processing on the vibration signal matrix, and/or After the pulse wave signal is obtained, the method further comprises inputting the pulse wave signal into a preset second filter so as to carry out filtering processing on the pulse wave signal.
5. The method for extracting pulse wave signals according to claim 1, wherein the predetermined space vector decomposition model includes a first space matrix and a second space matrix, wherein the pulse wave signal space matrix is obtained by the following formula: The pulse wave signal space matrix=the first space matrix (the second space matrix is the time-series-based vibration signal), The shape of the first space matrix is 1*3, the shape of the second space matrix is 3×4, the shape of the vibration signal based on the time sequence is 4×n, and n is the number of video images.
6. The method for extracting pulse wave signals according to claim 1, wherein the pulse wave signal space matrix is normalized by the following formula: , Wherein, the For the kth element in the pulse wave signal space matrix, n is the number of elements in the pulse wave signal space matrix, alpha is an empirical coefficient, The kth element of the pulse wave signal obtained after normalization processing.
7. An extraction device of pulse wave signals, characterized by being adapted to implement the method of any one of claims 1-6, comprising: the acquisition module is used for acquiring video data of a detected target, wherein the video data comprises a plurality of frames of video images and time stamps corresponding to the video images; the determining module is used for determining the information value of each frame of video image and obtaining a vibration signal based on a time sequence according to the information value and the corresponding time stamp; The generation module is used for inputting the vibration signals based on the time sequence into a preset space vector decomposition model and outputting a pulse wave signal space matrix; And the processing module is used for carrying out normalization processing on the pulse wave signal space matrix to obtain the pulse wave signal of the tested target.
8. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method of extracting pulse wave signals according to any one of claims 1-6.
9. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program, characterized in that the computer program, when executed by the processor, implements the method of extracting pulse wave signals according to any of claims 1-6.

Description

Pulse wave signal extraction method, pulse wave signal extraction device, storage medium and electronic equipment Technical Field The invention relates to the technical fields of bioengineering and artificial intelligence, in particular to a pulse wave signal extraction method, a pulse wave signal extraction device, a computer readable storage medium and electronic equipment. Background In the related art, when a pulse wave signal is obtained through a video (image sequence), noise outside an effective frequency band (1 hz-3.33 hz) is removed mainly through a frequency domain filtering means to obtain the pulse wave signal, and a small number of signals related to heart rate are obtained through detecting the capacity of hemoglobin. However, the signal obtained by the above technique is not clean because noise in the frequency band is not removed, and the signal obtained by hemoglobin has a small correlation with the pulse wave signal. The noise has a large duty cycle in the signal and may even completely drown the pulse wave signal. Disclosure of Invention The present invention aims to solve at least one of the technical problems in the related art to some extent. Therefore, a first object of the present invention is to provide a method for extracting pulse wave signals, which can obtain pulse wave signals of a pure target object comparable to a contact pulse wave sensor. A second object of the present invention is to provide an apparatus for extracting pulse wave signals. A third object of the present invention is to propose a computer readable storage medium. A fourth object of the present invention is to propose an electronic device. In order to achieve the above purpose, an embodiment of the first aspect of the present invention provides a method for extracting pulse wave signals, which includes the steps of collecting video data of a measured object, wherein the video data includes a plurality of frames of video images and time stamps corresponding to the video images, determining information values of each frame of video images, obtaining vibration signals based on time sequences according to the information values and the corresponding time stamps, inputting the vibration signals based on the time sequences into a preset space vector decomposition model, outputting a pulse wave signal space matrix, and normalizing the pulse wave signal space matrix to obtain pulse wave signals of the measured object. According to one embodiment of the invention, determining the information value of the video image comprises preprocessing the video image to obtain a region of interest of the video image, carrying out RGB color vector decomposition on the region of interest to obtain an RGB color matrix, and calculating an average value of each color component in the RGB color matrix to obtain a corresponding information value. According to one embodiment of the invention, the method for obtaining the vibration signal based on the time sequence according to the information values and the corresponding time stamps comprises the steps of obtaining a vibration signal matrix by combining the information values of the video images according to the time sequence, obtaining a time stamp matrix by combining the time stamps of the video images according to the time sequence, and obtaining the vibration signal based on the time sequence by combining the vibration signal matrix and the time stamp matrix. Further, after the vibration signal matrix is obtained, the method further comprises inputting the vibration signal matrix into a preset first filter to perform filtering processing on the vibration signal matrix, and/or after the pulse wave signal is obtained, inputting the pulse wave signal into a preset second filter to perform filtering processing on the pulse wave signal. According to one embodiment of the present invention, the preset space vector decomposition model includes a first space matrix and a second space matrix, wherein the pulse wave signal space matrix is obtained by the following formula that the pulse wave signal space matrix=the first space matrix (the second space matrix is the vibration signal based on time sequence), wherein the shape of the first space matrix is 1*3, the shape of the second space matrix is 3×4, the shape of the vibration signal based on time sequence is 4×n, and n is the number of video images. According to one embodiment of the invention, the method further comprises performing a frame interpolation process on the time-series based vibration signal according to the time stamp before inputting the time-series based vibration signal to a preset space vector decomposition model. Further, the frame inserting processing is carried out on the vibration signal based on the time sequence according to the time stamp, the frame inserting processing comprises the steps of calculating the difference value of time stamps of any two adjacent information values in the vibration signal based on the ti