A method and device for detecting pulse wave velocity based on video images

Abstract

The present invention discloses a pulse wave velocity detection method and device based on video images, the method comprising: obtaining video data of a preset body part of a person to be detected in a static state for a preset time; processing the video data, obtaining all frame images, segmenting each frame image respectively, extracting a first target area and a second target area in each frame image; converting each frame image into a grayscale format, calculating the actual distance between the grayscale center of gravity of the first target area and the grayscale center of gravity of the second target area; extracting the pulse wave signal corresponding to the first target area and the second target area, using a differential method to calculate the time difference between the pulse wave signal of the first target area and the pulse wave signal of the second target area; and calculating the pulse wave velocity based on the above processing results. The present invention does not require a complex detection device, reducing the workload and harm to the human body.

Description

Technical Field

[0001] The present invention relates to the technical field of non-contact physiological parameter detection, and in particular to a pulse wave velocity detection method and device based on video images. Background Art

[0002] Pulse wave refers to the process in which the heart beats and ejects blood into the aorta in a fluctuating manner. The aortic wall therefore generates a pulse pressure wave, referred to as pulse wave. The pulse wave is conducted along the blood vessel wall to the peripheral blood vessels at a certain speed. This speed is called pulse wave velocity (PWV). Pulse wave velocity is an important criterion for evaluating arteriosclerosis. The detection of pulse wave velocity can greatly help clinicians discover the degree of vascular damage in hypertensive patients; pulse wave velocity is of great significance in the assessment of the risk and prognosis of cardiovascular and cerebrovascular diseases. Clinically, the pulse wave velocity is generally calculated by measuring and recording the pulse wave conduction time and distance between two arterial sites. For example, the brachial artery and ankle artery are selected to measure the brachial-ankle PWV, the carotid artery and brachial artery are selected to measure the upper arm PWV, and the carotid artery and femoral artery are selected to measure the carotid-femoral PWV.

[0003] Planar tension method is one of the traditional methods for measuring PWV, which is mainly applicable to superficial arteries, such as radial artery, femoral artery and carotid artery. The measurement process includes the following steps: 1. Select the measurement site and measure the distance between the two points; 2. Place the pressure receptor at the most obvious pulsation point of the measurement site; 3. Start the PWV measurement device. There are several factors that need to be paid attention to in the plane tension method. First, the selection of the measurement position of the sensor should reduce the influence of the activities of the operator and the subject; second, it is necessary to make the probe as perpendicular to the axis of the blood vessel as possible; third, it is necessary to make sure that the force of pressing down can just flatten the artery. Another method to measure PWV is the catheter method, which is an invasive method. The angiography catheter is inserted into the arterial sheath, and the catheter head is sent to the aortic root. A pressure transducer is connected to the tail end of the catheter and the side hole of the arterial sheath respectively. The pressure transducer is connected to the physiological recorder to obtain the intra-arterial pressure waveform at the aortic root and the iliac artery. The distance between the starting points of the two pressure waveforms is the time difference T of the pulse wave conduction. 1 ; Then measure the length of the angiography catheter entering the body, and subtract the length of the arterial sheath to get the distance L between the two points. 1 Then withdraw the catheter tip to the abdominal aorta and repeat the above measurement to obtain T 2 and L 2 , the distance between the aortic root and the abdominal aorta is L 1 -L 2 , the transmission time of the pulse wave between these two points is T 1 -T 2, the pulse wave transmission distance divided by the pulse wave transmission time can get PWV. Although the catheter method has high measurement accuracy, it is not accepted by most patients because it will cause certain harm to the patient. Therefore, the catheter method has not been popularized in clinical practice. In recent years, the oscillometric method is the most commonly used method for measuring PWV in clinical practice. By using oscillometric measurement technology, the PWV of the arm ankle is measured to achieve automation of PWV measurement and improve measurement efficiency. The advantages of the oscillometric method are simple operation, high repeatability and high accuracy. However, the existing detection methods require the use of complex detection devices and have a large workload. Summary of the invention

[0004] The present invention provides a method and device for detecting pulse wave velocity based on video images, so as to solve the technical problems existing in the prior art of complex detection devices, large workload and harm to the human body.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In one aspect, the present invention provides a method for detecting pulse wave velocity based on video images, the method comprising:

[0007] Acquire video data of a preset body part of a person to be tested in a static state for a preset time period;

[0008] Processing the video data, acquiring all frame images in the video data, and segmenting each acquired frame image respectively, extracting a first target area and a second target area in each frame image; wherein, in each frame image, the position and size of the first target area in the image are the same; and at the same time, in each frame image, the position and size of the second target area in the image are also the same;

[0009] Convert each frame of the image into a grayscale format, obtain the grayscale centroids of the first target area and the second target area, and calculate the actual distance between the grayscale centroids of the first target area and the second target area; extract the pulse wave signals corresponding to the first target area and the second target area, and use the difference method to calculate the time difference between the pulse wave signal of the first target area and the pulse wave signal of the second target area;

[0010] The actual distance between the two grayscale centers of gravity is used as the conduction distance of the pulse wave, and the time difference between the two pulse wave signals is used as the conduction time of the pulse wave, and the conduction velocity of the pulse wave is calculated.

[0011] Further, obtaining the grayscale centroids of the first target area and the second target area includes:

[0012] The coordinate value of the grayscale centroid of the corresponding target area in the single-frame image is calculated by the following formula; wherein the corresponding target area refers to the first target area or the second target area;

[0013]

[0014]

[0015] Among them, a 0 , b 0 Respectively represent the horizontal and vertical coordinates of the grayscale centroid of the corresponding target area; Q represents the corresponding target area; W(x, y) represents the grayscale value of the point with coordinates (x, y) in the corresponding target area;

[0016] The coordinates of the grayscale centroids of all frame images are averaged to obtain the grayscale centroid position of the corresponding target area.

[0017] Further, the calculating the actual distance between the grayscale centroid of the first target area and the grayscale centroid of the second target area includes:

[0018] Based on the grayscale centroid position of the first target area and the grayscale centroid position of the second target area, a distance between the grayscale centroid of the first target area and the grayscale centroid of the second target area in the image is calculated;

[0019] The scale of the distance in the image and the actual distance is obtained, and the actual distance between the two grayscale centroids is calculated based on the scale and the distance between the two grayscale centroids in the image according to the ratio conversion.

[0020] Further, extracting the pulse wave signals corresponding to the first target area and the second target area includes:

[0021] Calculate the third quartile of all pixels in the target area corresponding to each frame of image, and obtain the mean of the third quartile of pixels corresponding to all frames of image; for each frame of image, count the total number of pixels in the corresponding target area whose grayscale value is greater than the mean, and use the counted total number of pixels as the intensity of the pulse wave signal at the moment corresponding to the current frame of image; draw a curve with the number of image frames as the horizontal coordinate and the intensity of the pulse wave signal corresponding to each frame of image as the vertical coordinate, as the original pulse wave signal corresponding to the corresponding target area; wherein the corresponding target area refers to the first target area or the second target area;

[0022] Performing empirical mode decomposition on the original pulse wave signal to decompose signal components of different frequency components, filtering out low-frequency baseline shift and high-frequency noise interference components according to the frequency characteristics of the pulse wave signal to obtain a filtered pulse wave signal, and using the filtered pulse wave signal as the final pulse wave signal.

[0023] Furthermore, the step of calculating the time difference between the pulse wave signal of the first target area and the pulse wave signal of the second target area by using a differential method includes:

[0024] All peaks and troughs of the filtered pulse wave signals corresponding to the first target area and the second target area are obtained respectively; the average value of the adjacent peak time frame difference and the average value of the adjacent trough time frame difference of the two pulse wave signals are calculated, and then the average value of the peak time frame difference and the average value of the trough time frame difference are averaged, and the calculation result is recorded as the time difference of the pulse wave signals of the two target areas.

[0025] On the other hand, the present invention also provides a pulse wave velocity detection device based on video images, the pulse wave velocity detection device based on video images comprising:

[0026] A video data acquisition module, used to acquire video data of a preset body part of a subject to be detected in a static state for a preset time period;

[0027] The data processing module is used to perform the following steps:

[0028] Processing the video data, acquiring all frame images in the video data, and segmenting each acquired frame image respectively, extracting a first target area and a second target area in each frame image; wherein, in each frame image, the position and size of the first target area in the image are the same; and at the same time, in each frame image, the position and size of the second target area in the image are also the same;

[0029] Convert each frame of the image into a grayscale format, obtain the grayscale centroids of the first target area and the second target area, and calculate the actual distance between the grayscale centroids of the first target area and the second target area; extract the pulse wave signals corresponding to the first target area and the second target area, and use the difference method to calculate the time difference between the pulse wave signal of the first target area and the pulse wave signal of the second target area;

[0030] The actual distance between the two grayscale centers of gravity is used as the conduction distance of the pulse wave, and the time difference between the two pulse wave signals is used as the conduction time of the pulse wave, and the conduction velocity of the pulse wave is calculated.

[0031] Furthermore, the data processing module is specifically used for:

[0032] The coordinate value of the grayscale centroid of the corresponding target area in the single-frame image is calculated by the following formula; wherein the corresponding target area refers to the first target area or the second target area;

[0033]

[0034]

[0035] Among them, a 0 , b 0 Respectively represent the horizontal and vertical coordinates of the grayscale centroid of the corresponding target area; Q represents the corresponding target area; W(x, y) represents the grayscale value of the point with coordinates (x, y) in the corresponding target area;

[0036] The coordinates of the grayscale centroids of all frame images are averaged to obtain the grayscale centroid position of the corresponding target area.

[0037] Furthermore, the data processing module is also specifically used for:

[0038] Based on the grayscale centroid position of the first target area and the grayscale centroid position of the second target area, a distance between the grayscale centroid of the first target area and the grayscale centroid of the second target area in the image is calculated;

[0039] The scale of the distance in the image and the actual distance is obtained, and the actual distance between the two grayscale centroids is calculated based on the scale and the distance between the two grayscale centroids in the image according to the ratio conversion.

[0040] Furthermore, the data processing module is also specifically used for:

[0041] Calculate the third quartile of all pixels in the target area corresponding to each frame of image, and obtain the mean of the third quartile of pixels corresponding to all frames of image; for each frame of image, count the total number of pixels in the corresponding target area whose grayscale value is greater than the mean, and use the counted total number of pixels as the intensity of the pulse wave signal at the moment corresponding to the current frame of image; draw a curve with the number of image frames as the horizontal coordinate and the intensity of the pulse wave signal corresponding to each frame of image as the vertical coordinate, as the original pulse wave signal corresponding to the corresponding target area; wherein the corresponding target area refers to the first target area or the second target area;

[0042] Performing empirical mode decomposition on the original pulse wave signal to decompose signal components of different frequency components, filtering out low-frequency baseline shift and high-frequency noise interference components according to the frequency characteristics of the pulse wave signal to obtain a filtered pulse wave signal, and using the filtered pulse wave signal as the final pulse wave signal.

[0043] Furthermore, the data processing module is also specifically used for:

[0044] All peaks and troughs of the filtered pulse wave signals corresponding to the first target area and the second target area are obtained respectively; the average value of the adjacent peak time frame difference and the average value of the adjacent trough time frame difference of the two pulse wave signals are calculated, and then the average value of the peak time frame difference and the average value of the trough time frame difference are averaged, and the calculation result is recorded as the time difference of the pulse wave signals of the two target areas.

[0045] On the other hand, the present invention further provides an electronic device, comprising a processor and a memory; wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the above method.

[0046] In yet another aspect, the present invention further provides a computer-readable storage medium, wherein the storage medium stores at least one instruction, and the instruction is loaded and executed by a processor to implement the above method.

[0047] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0048] The pulse wave conduction velocity detection scheme provided by the present invention uses a smart phone or other video recording equipment to shoot 10s or 30s video image data of the wrist or arm of a person to be detected in a normal video recording state, uses a computer and corresponding software to process the collected data, obtains video image frame data of each frame of the video, divides the image frame to obtain two target areas, and counts the total number of target area grayscale values ​​greater than the third quartile of all image frames to obtain the original pulse wave signals of the two target areas; performs empirical mode decomposition on the signal to filter out baseline translation and high-frequency noise to obtain a relatively pure pulse wave signal, uses a difference method to calculate the positions of the peaks and troughs of the pulse waves of the two target areas, respectively calculates the adjacent peak time difference and the adjacent trough time difference of the pulse waves of the two target areas, counts the average value thereof and finally uses it as the waveform time difference of the pulse waves of the two target areas, uses a grayscale centroid method to calculate the grayscale centroid of the two target areas, measures the distance between the two centroids, and then converts according to the actual ratio to obtain the actual distance between the two grayscale centroids. Finally, according to the calculation formula of PWV, the distance between the grayscale centers of the two target areas is divided by the time difference of the pulse waves of the two target areas to calculate the pulse wave conduction velocity. The present invention uses a non-contact video method to obtain pulse wave signals, does not require complex detection devices, is easy to operate, saves time, and reduces workload and harm to the human body. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0050] Figure 1 1 is a schematic diagram of an execution flow of a pulse wave velocity detection method based on video images provided in an embodiment of the present invention;

[0051] Figure 2 is one frame of an image of a 10-second arm video taken by a mobile phone provided in an embodiment of the present invention;

[0052] Figure 3 The embodiment of the present invention provides Figure 2 The image frame is segmented to obtain the image of the target area;

[0053] Figure 4 It is a schematic diagram of the execution flow of a method for extracting an original pulse wave signal of a target area and calculating a time difference between pulse wave signals of two target areas provided by an embodiment of the present invention;

[0054] Figure 5 is a graph of original pulse wave signals extracted from a target area above all image frames provided by an embodiment of the present invention;

[0055] Fig. 6A The embodiment of the present invention provides Figure 5 Schematic diagram of various signal components obtained by performing empirical mode decomposition of the pulse wave signal shown;

[0056] Figure 6B The embodiment of the present invention provides Figure 5 The graph of the relatively pure pulse wave signal obtained after the pulse wave signal is denoised is shown;

[0057] Figure 7 is a schematic diagram of an execution flow of a method for obtaining the grayscale centroids of two target areas and calculating the actual distance between the grayscale centroids of the two target areas provided by an embodiment of the present invention;

[0058] Figure 8 It is a pulse wave curve diagram of the two target areas of the 10s video shot by the mobile phone provided in an embodiment of the present invention after the pulse waves are extracted and filtered; wherein data1, that is, the solid line curve, represents the pulse wave signal of the upper target area, and data2, that is, the dotted line curve, represents the pulse wave signal of the lower target area. DETAILED DESCRIPTION

[0059] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0060] First embodiment

[0061] The present embodiment provides a pulse wave velocity detection method based on video images. The method can be implemented by an electronic device. The pulse wave velocity detection method requires a smart phone or other video recording device. The basic principle of the method is to use a video recording device to capture the arm part for 10 seconds or 30 seconds in a normal recording state to collect video image data, use a computer and corresponding software to process the collected data, obtain the image frame data of each frame of the video, segment the image frame to obtain two target areas, count the total number of target area grayscale values ​​greater than the third quartile of all image frames to obtain the original pulse wave signals of the two target areas, perform empirical mode decomposition on the signal to filter out baseline shift and high-frequency noise to obtain a relatively pure pulse wave signal, calculate the waveform time difference, use the grayscale center of gravity method to calculate the grayscale center of gravity of the target area, measure the actual distance between the two centers of gravity, and then calculate the pulse wave velocity according to the PWV calculation formula. Specifically, the execution flow of the pulse wave velocity detection method based on video images is as follows: Figure 1 As shown, the following steps are included:

[0062] S1, obtaining video data of a preset body part of a person to be detected in a static state for a preset time period;

[0063] It should be noted that, in the present embodiment, the extraction of the pulse wave signal is derived from a video captured by a smart phone, camera or other video recording device; therefore, during the test, it is necessary to use a video recording device to capture 10 seconds or 30 seconds of video data of the arm of the person to be tested in a static state under normal working conditions, and use a connecting device to transfer it to the relevant computer software for storage, so that the corresponding software can be used to process the video data later.

[0064] Of course, it is understandable that, firstly, the device for shooting video is not limited to smart phones, and other devices with shooting functions such as cameras, monitoring and the like can be used as shooting devices of the present invention to obtain original videos, and in order to ensure that sufficient pulse wave information can be obtained from the video, the resolution of the video image should not be less than 1920*1080; secondly, the part of the person to be tested that is shot is not limited to the arm of the person to be tested, and the limbs and even the face of the person to be tested can also be used as the video source for obtaining the pulse wave signal of the present invention. Compared with other parts of the body, the arm has the advantages of rich blood vessels, strong pulse wave signals, and straight arteries; finally, when shooting the video, try to keep the shooting part still and reduce jitter, and the length of the video shooting is not limited to 10s and 30s, 10s to 30s is a more suitable time interval.

[0065] S2, processing the video data, acquiring all frame images in the video data, and segmenting each acquired frame image respectively, extracting a first target area and a second target area in each frame image; wherein, in each frame image, the position and size of the first target area in the image are the same; and at the same time, in each frame image, the position and size of the second target area in the image are also the same;

[0066] Specifically, in this embodiment, the duration of the captured video is 10.56s, the video frame rate is 30.0583, the total number of video frames is 317, the video resolution is 1920*1080, the video format is RGB24, and one frame of the video is as follows: Figure 2 As shown. Extract all the frame images of the video, segment each image frame to obtain two target areas. Figure 2 The results of the image frame segmentation shown in Figure 3 As shown in the figure, the areas encircled by the two black boxes are the target areas (upper target area and lower target area). The size of the upper black box is 301*301, and the size of the lower black box is 401*401. It should be pointed out that the selection of the target area is not limited to the above area sizes, but should be determined according to different shooting parts and different situations. The selection principle is to include blood vessels and arteries as much as possible. In addition, it should also be adjusted according to the differences in different limbs of different people.

[0067] S3, converting each frame image into a grayscale format, obtaining the grayscale centroids of the first target area and the second target area, and calculating the actual distance between the grayscale centroids of the first target area and the second target area; extracting the pulse wave signals corresponding to the first target area and the second target area, and using a differential method to calculate the time difference between the pulse wave signals of the first target area and the second target area;

[0068] Among them, in this embodiment, in order to facilitate the subsequent calculation of the grayscale centroid, the image format of each frame is converted from RGB format to grayscale format. The above S3 first obtains the original pulse wave curve by counting the grayscale features of the pixels of the two target areas in all frames of the video, and then performs empirical mode decomposition on the original pulse wave to remove the baseline translation and high-frequency noise to obtain a relatively pure pulse wave signal, and based on this, the time difference of the pulse wave signals of the two target areas is calculated using the difference method. And the above S3 also obtains the grayscale centroid of the two target areas, calculates the distance between the grayscale centroids of the two target areas, and measures the actual distance.

[0069] Specifically, in this embodiment, the execution flow of the method for extracting the original pulse wave signal of the target area and calculating the time difference between the pulse wave signals of the two target areas is as follows: Figure 4 As shown, the following steps are included:

[0070] Step 1: Count the grayscale features of the two target areas respectively, and calculate the original pulse wave signals of the two target areas based on the grayscale features counted, as follows:

[0071] Calculate the third quartile of all pixels in the target area of ​​each frame image, and obtain the mean of the third quartile of all frame pixels. For each frame image, count the total number of pixels in the target area whose grayscale value is greater than the mean. The total number of pixels is used as the intensity of the pulse wave signal at the moment of the image frame. Plot a curve between this value of all image frames and the number of image frames to obtain the pulse wave signal curve, such as Figure 5 As shown, the horizontal axis F represents the number of image frames, and the vertical axis S represents the acquired pulse wave signal strength. Figure 5 The pulse wave shown contains a lot of noise interference, and it can be clearly seen that there is interference from baseline shift.

[0072] Step 2: Perform empirical mode decomposition on the original pulse wave signal to decompose signal components of different frequency components, filter out low-frequency baseline shift and high-frequency noise interference components according to the frequency characteristics of the pulse wave signal, and use the relatively pure pulse wave signal obtained after filtering as the final pulse wave signal.

[0073] Among them, it should be noted that the empirical mode decomposition of the pulse wave signal is to obtain the intrinsic mode function. According to the concept of intrinsic mode function, any signal is composed of several intrinsic mode functions, and the intrinsic mode functions overlap and affect each other to form a composite function. Therefore, the empirical mode decomposition of the pulse wave signal is to obtain the intrinsic mode functions of each order of the pulse wave signal. The specific steps of empirical mode decomposition of the pulse wave signal are as follows. The first step is to obtain all the maximum points of the original pulse wave signal, and use the cubic spline function to fit the upper envelope. The second step is to obtain all the minimum points of the original pulse wave signal, and use the cubic spline function to fit the lower envelope. The third step is to record the mean of the upper envelope and the lower envelope as the average envelope. The original pulse wave signal is subtracted from the average envelope to obtain a new signal. This signal is an intrinsic mode function of the original pulse wave. The fourth step is to obtain the first-order or higher-order differential of the original pulse wave, and the obtained signal is repeated from steps one to three, thereby obtaining the intrinsic mode function of the original pulse wave. Figure 5 The results of the empirical mode decomposition of the original pulse wave are shown as Fig. 6A As shown, according to the frequency characteristics of the pulse wave signal, the signal components of imf1 and imf2 belong to low-frequency noise interference, the signal components of imf5 and imf6 belong to high-frequency noise, and res is the residual interference caused by baseline translation. Therefore, the sum of the signal components of imf3 and imf4 is the extracted pulse wave signal component, as shown in Figure 6B Shown is Figure 5The pulse wave shown is a relatively pure pulse wave signal obtained after the pulse wave is filtered by empirical mode decomposition. Among them, the horizontal axis F represents the number of image frames, and the vertical axis S represents the intensity of the acquired pulse wave signal.

[0074] The pulse wave curves after the pulse wave filtering process of the two target areas extracted from the 10s video shot by the mobile phone are shown as follows: Figure 8 As shown; wherein data1, that is, the solid line curve, represents the pulse wave signal of the upper target area, and data2, that is, the dotted line curve, represents the pulse wave signal of the lower target area.

[0075] Step 3: Use the differential method to calculate the time difference between the pulse wave signals of the two target areas, as follows:

[0076] The first-order difference of the pulse wave after filtering is performed, that is, the difference of the pulse wave signals of two adjacent frames is performed. At the peak, the adjacent differential value in front is a positive number, and the adjacent differential value behind is a negative number. Based on this characteristic, all the peaks of the pulse wave can be obtained. However, since there are small peaks in the pulse wave that are not the true peaks of the pulse wave waveform, this embodiment sets 1.2 times the average value of the pulse wave intensity as the threshold. If the detected peak is less than this value, the peak will be discarded; at the trough, the adjacent differential value in front is a negative number, and the adjacent differential value behind is a positive number. With the above method for obtaining the peak, all the troughs of the pulse wave can also be obtained. After obtaining all the peaks and troughs of the filtered pulse wave signals corresponding to the two target areas respectively, first calculate the average value of the adjacent peak time frame difference and the average value of the adjacent trough time frame difference of the two pulse waves, and then average the average value of the peak time frame difference and the average value of the trough time frame difference, and the result is recorded as the waveform time difference of the pulse waves of the two target areas.

[0077] Further, in this embodiment, the execution flow of the method for obtaining the grayscale centroids of two target areas and calculating the actual distance between the grayscale centroids of the two target areas is as follows: Figure 7 As shown, the following steps are included:

[0078] Step 1: Get the grayscale centroid positions of the two target areas, as follows:

[0079] For a frame of video, the square of the grayscale value of all pixels in the target area is used as the weight, and the horizontal and vertical coordinates of all pixels are multiplied by the weight, and then the weight is averaged to obtain the horizontal and vertical coordinates of the grayscale center of gravity. The formula is as follows:

[0080]

[0081]

[0082] Among them, a 0 、b0 Respectively represent the horizontal and vertical coordinates of the grayscale centroid of the corresponding target area; Q represents the corresponding target area; W(x, y) represents the grayscale value of the point with coordinates (x, y) in the corresponding target area;

[0083] The coordinates of the grayscale centroids of all frame images are averaged to obtain the grayscale centroid position of the corresponding target area.

[0084] Step 2: Calculate the distance between the grayscale centroids of the two target areas in the image according to the distance formula.

[0085] Step 3, obtaining a scale between the distance in the image and the actual distance, and calculating the actual distance between the two grayscale centroids according to the scale and the distance between the two grayscale centroids in the image according to the ratio conversion.

[0086] Specifically, this embodiment records the arm positions corresponding to the upper and lower boundaries in the shot video, measures the actual distances between the upper and lower boundaries of the shot arm video, and obtains the ratio of the shot video to the actual situation based on the pixel distances of the upper and lower boundaries. Then, based on the ratio of the shot video to the actual situation and the distance between the two grayscale centers of gravity in the image, the actual distance between the two grayscale centers of gravity is converted.

[0087] S4, taking the actual distance between the two grayscale gravity centers as the conduction distance of the pulse wave, taking the time difference between the two pulse wave signals as the conduction time of the pulse wave, and calculating the conduction velocity of the pulse wave.

[0088] Specifically, this embodiment obtains the pulse wave conduction velocity of the arm video by dividing the actual distance of the grayscale center of gravity calculated by S3 by the pulse wave time difference calculated by S3.

[0089] In summary, the pulse wave velocity detection method of the present embodiment uses a smart phone or other video recording equipment to shoot 10s or 30s video image data of the wrist or arm of the person to be detected in a normal recording state, uses a computer and corresponding software to process the collected data, obtains each frame of the video image frame data, divides the image frame to obtain two target areas, and counts the total number of target area grayscale values ​​greater than the third quartile of all image frames to obtain the original pulse wave signals of the two target areas; performs empirical mode decomposition on the signal to filter out baseline translation and high-frequency noise to obtain a relatively pure pulse wave signal, uses the difference method to calculate the positions of the peaks and troughs of the pulse waves of the two target areas, calculates the adjacent peak time difference and the adjacent trough time difference of the pulse waves of the two target areas respectively, and finally counts the mean value as the waveform time difference of the pulse waves of the two target areas, calculates the grayscale center of gravity of the two target areas using the grayscale center of gravity method, measures the distance between the two centers of gravity, and then converts according to the actual ratio to obtain the actual distance between the two grayscale centers of gravity. Finally, according to the calculation formula of PWV, the distance between the grayscale centers of the two target areas is divided by the time difference of the pulse waves of the two target areas to calculate the pulse wave conduction velocity. The present invention uses a non-contact video method to obtain pulse wave signals, does not require complex detection devices, is easy to operate, saves time, and reduces workload and harm to the human body.

[0090] Second embodiment

[0091] This embodiment provides a pulse wave velocity detection device based on video images, comprising:

[0092] A video data acquisition module, used to acquire video data of a preset body part of a subject to be detected in a static state for a preset time period;

[0093] The data processing module is used to perform the following steps:

[0094] Processing the video data, acquiring all frame images in the video data, and segmenting each acquired frame image respectively, extracting a first target area and a second target area in each frame image; wherein, in each frame image, the position and size of the first target area in the image are the same; and at the same time, in each frame image, the position and size of the second target area in the image are also the same;

[0095] Convert each frame of the image into a grayscale format, obtain the grayscale centroids of the first target area and the second target area, and calculate the actual distance between the grayscale centroids of the first target area and the second target area; extract the pulse wave signals corresponding to the first target area and the second target area, and use the difference method to calculate the time difference between the pulse wave signal of the first target area and the pulse wave signal of the second target area;

[0096] The actual distance between the two grayscale centers of gravity is used as the conduction distance of the pulse wave, and the time difference between the two pulse wave signals is used as the conduction time of the pulse wave, and the conduction velocity of the pulse wave is calculated.

[0097] The pulse wave velocity detection device based on video images of the present embodiment corresponds to the pulse wave velocity detection method based on video images of the first embodiment; the pulse wave velocity detection device based on video images of the present embodiment is used to implement the pulse wave velocity detection method based on video images of the first embodiment; wherein, the functions implemented by each functional module in the pulse wave velocity detection device based on video images of the present embodiment correspond one-to-one to each process step in the pulse wave velocity detection method based on video images of the first embodiment; therefore, they will not be described in detail here.

[0098] Third embodiment

[0099] This embodiment provides an electronic device, which includes a processor and a memory; wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method of the first embodiment.

[0100] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) and one or more memories, wherein the memory stores at least one instruction, and the instruction is loaded by the processor to execute the above method.

[0101] Fourth embodiment

[0102] This embodiment provides a computer-readable storage medium, which stores at least one instruction, and the instruction is loaded and executed by a processor to implement the method of the first embodiment. The computer-readable storage medium may be a ROM, a random access memory, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc. The instructions stored therein may be loaded by a processor in a terminal to execute the method.

[0103] In addition, it should be noted that the present invention can be provided as a method, an apparatus or a computer program product. Therefore, the embodiments of the present invention can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the embodiments of the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code.

[0104] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0105] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0106] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0107] Finally, it should be noted that the above is a preferred embodiment of the present invention. It should be pointed out that although the preferred embodiment of the present invention has been described, for those skilled in the art, once the basic creative concept of the present invention is known, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention. Therefore, the attached claims are intended to be interpreted as including the preferred embodiment and all changes and modifications that fall within the scope of the embodiments of the present invention.

Claims

1. A pulse wave velocity detection method based on video images, It is characterized in that include: Acquire video data of a preset body part of a person to be tested in a static state for a preset time period; Processing the video data, acquiring all frame images in the video data, and segmenting each acquired frame image respectively, extracting a first target area and a second target area in each frame image; wherein, in each frame image, the position and size of the first target area in the image are the same; and at the same time, in each frame image, the position and size of the second target area in the image are also the same; Convert each frame of the image into a grayscale format, obtain the grayscale centroids of the first target area and the second target area, and calculate the actual distance between the grayscale centroids of the first target area and the second target area; extract the pulse wave signals corresponding to the first target area and the second target area, and use the difference method to calculate the time difference between the pulse wave signal of the first target area and the pulse wave signal of the second target area; The actual distance between the two grayscale centers of gravity is used as the conduction distance of the pulse wave, and the time difference between the two pulse wave signals is used as the conduction time of the pulse wave, and the conduction velocity of the pulse wave is calculated; Obtaining the grayscale centroid of the first target area and the second target area includes: The coordinate value of the grayscale centroid of the corresponding target area in the single-frame image is calculated by the following formula; wherein the corresponding target area refers to the first target area or the second target area; Among them, a 0 、b 0 Respectively represent the horizontal and vertical coordinates of the grayscale centroid of the corresponding target area; Q represents the corresponding target area; W(x, y) represents the grayscale value of the point with coordinates (x, y) in the corresponding target area; The coordinates of the grayscale centroids of all frame images are averaged to obtain the grayscale centroid position of the corresponding target area; The calculating the actual distance between the grayscale centroid of the first target area and the grayscale centroid of the second target area includes: Based on the grayscale centroid position of the first target area and the grayscale centroid position of the second target area, a distance between the grayscale centroid of the first target area and the grayscale centroid of the second target area in the image is calculated; Obtaining a scale between the distance in the image and the actual distance, and calculating the actual distance between the two grayscale centroids according to the scale and the distance between the two grayscale centroids in the image according to the ratio conversion; Extracting the pulse wave signals corresponding to the first target area and the second target area includes: Calculate the third quartile of all pixels in the target area corresponding to each frame of image, and obtain the mean of the third quartile of pixels corresponding to all frames of image; for each frame of image, count the total number of pixels in the corresponding target area whose grayscale value is greater than the mean, and use the counted total number of pixels as the intensity of the pulse wave signal at the moment corresponding to the current frame of image; draw a curve with the number of image frames as the horizontal coordinate and the intensity of the pulse wave signal corresponding to each frame of image as the vertical coordinate, as the original pulse wave signal corresponding to the corresponding target area; wherein the corresponding target area refers to the first target area or the second target area; Performing empirical mode decomposition on the original pulse wave signal to decompose signal components of different frequency components, filtering out low-frequency baseline shift and high-frequency noise interference components according to the frequency characteristics of the pulse wave signal to obtain a filtered pulse wave signal, and using the filtered pulse wave signal as the final pulse wave signal.

2. The pulse wave velocity detection method based on video images as claimed in claim 1, It is characterized in that The method of using a differential method to calculate the time difference between the pulse wave signal of the first target area and the pulse wave signal of the second target area includes: All peaks and troughs of the filtered pulse wave signals corresponding to the first target area and the second target area are obtained respectively; the average value of the adjacent peak time frame difference and the average value of the adjacent trough time frame difference of the two pulse wave signals are calculated, and then the average value of the peak time frame difference and the average value of the trough time frame difference are averaged, and the calculation result is recorded as the time difference of the pulse wave signals of the two target areas.

3. A pulse wave velocity detection device based on video images, It is characterized in that include: A video data acquisition module, used to acquire video data of a preset body part of a subject to be detected in a static state for a preset time period; The data processing module is used to perform the following steps: Processing the video data, acquiring all frame images in the video data, and segmenting each acquired frame image respectively, extracting a first target area and a second target area in each frame image; wherein, in each frame image, the position and size of the first target area in the image are the same; and at the same time, in each frame image, the position and size of the second target area in the image are also the same; Convert each frame of the image into a grayscale format, obtain the grayscale centroids of the first target area and the second target area, and calculate the actual distance between the grayscale centroids of the first target area and the second target area; extract the pulse wave signals corresponding to the first target area and the second target area, and use the difference method to calculate the time difference between the pulse wave signal of the first target area and the pulse wave signal of the second target area; The actual distance between the two grayscale centers of gravity is used as the conduction distance of the pulse wave, and the time difference between the two pulse wave signals is used as the conduction time of the pulse wave, and the conduction velocity of the pulse wave is calculated; The data processing module is specifically used for: The coordinate value of the grayscale centroid of the corresponding target area in the single-frame image is calculated by the following formula; wherein the corresponding target area refers to the first target area or the second target area; Among them, a 0 , b 0 Respectively represent the horizontal and vertical coordinates of the grayscale centroid of the corresponding target area; Q represents the corresponding target area; W(x, y) represents the grayscale value of the point with coordinates (x, y) in the corresponding target area; The coordinates of the grayscale centroids of all frame images are averaged to obtain the grayscale centroid position of the corresponding target area; The data processing module is also specifically used for: Based on the grayscale centroid position of the first target area and the grayscale centroid position of the second target area, a distance between the grayscale centroid of the first target area and the grayscale centroid of the second target area in the image is calculated; Obtaining a scale between the distance in the image and the actual distance, and calculating the actual distance between the two grayscale centroids according to the scale and the distance between the two grayscale centroids in the image according to the ratio conversion; The data processing module is also specifically used for: Calculate the third quartile of all pixels in the target area corresponding to each frame of image, and obtain the mean of the third quartile of pixels corresponding to all frames of image; for each frame of image, count the total number of pixels in the corresponding target area whose grayscale value is greater than the mean, and use the counted total number of pixels as the intensity of the pulse wave signal at the moment corresponding to the current frame of image; draw a curve with the number of image frames as the horizontal coordinate and the intensity of the pulse wave signal corresponding to each frame of image as the vertical coordinate, as the original pulse wave signal corresponding to the corresponding target area; wherein the corresponding target area refers to the first target area or the second target area; Performing empirical mode decomposition on the original pulse wave signal to decompose signal components of different frequency components, filtering out low-frequency baseline shift and high-frequency noise interference components according to the frequency characteristics of the pulse wave signal to obtain a filtered pulse wave signal, and using the filtered pulse wave signal as the final pulse wave signal.

4. The pulse wave velocity detection device based on video images as claimed in claim 3, It is characterized in that The data processing module is also specifically used for: All peaks and troughs of the filtered pulse wave signals corresponding to the first target area and the second target area are obtained respectively; the average value of the adjacent peak time frame difference and the average value of the adjacent trough time frame difference of the two pulse wave signals are calculated, and then the average value of the peak time frame difference and the average value of the trough time frame difference are averaged, and the calculation result is recorded as the time difference of the pulse wave signals of the two target areas.

Images