A method and device for measuring respiratory rate

Abstract

The present invention discloses a respiratory rate measurement method and device. The implementation of the method includes: collecting pressure map data and determining the chest and abdomen regions in the pressure map data; determining a target unoccupied area according to a preset rule, and obtaining first pressure data of the chest and abdomen regions and second pressure data of the target unoccupied area; performing collaborative filtering processing on the first pressure data by using the second pressure data to obtain an adaptive filtering signal; performing band-pass filtering processing on the adaptive filtering signal to obtain a respiratory wave signal; extracting peak points from the respiratory wave signal to obtain peak intervals; and calculating a respiratory rate value based on the peak intervals. By extracting the positions of the chest and abdomen and using the pressure signals of the unoccupied area to perform collaborative filtering on the pressure signals of the positions of the chest and abdomen, the signal quality is improved to a certain extent, the accuracy of the respiratory rate is increased, and the respiratory rate measurement can be realized in complex scenarios such as multiple people and multiple people and multiple objects. The applicable scenarios are extensive and the accuracy is higher.

Description

Technical Field

[0001] The present invention relates to the technical field of smart home, and particularly to a method and device for measuring respiratory rate, a computer device, and a storage medium. Background Art

[0002] With the development of the intelligence of mattresses, the functions of smart mattresses are becoming more and more powerful. Some mattresses have realized the monitoring of physiological parameter data (such as weight, heart rate, respiration, body movement, etc.). However, during the testing process, the measurement of respiratory rate is often affected by the environment, multiple people and objects, etc., resulting in unsatisfactory accuracy of respiratory rate measurement.

[0003] Currently, the existing respiratory rate measurement schemes are mainly divided into two types: contact type and non-contact type. Common contact type schemes include electrocardiogram, electromyogram, photoplethysmogram, etc., and non-contact type includes radar wave, image processing, etc. The existing respiratory rate measurement schemes based on the bed mainly include two types: based on ballistocardiogram (BCG) and airbag pressure signal. Such methods have achieved high accuracy in respiratory rate measurement in a single-person scenario with less environmental interference. However, when there are multiple people, there are other vibration sources in the environment, or there is a strong electromagnetic source in the environment, the measurement accuracy will be greatly reduced. Especially in the multi-person scenario, the measurement results of the existing methods are basically untrustworthy. Summary of the Invention

[0004] Based on this, it is necessary to provide a method and device for measuring respiratory rate, a computer device, and a storage medium for the above technical problems, so as to solve at least one of the problems existing in the above prior art.

[0005] In a first aspect, the embodiment of the present application is implemented as follows. A method for measuring respiratory rate is provided, including the following steps:

[0006] Collect pressure map data and determine the chest and abdomen area in the pressure map data;

[0007] Determine a target unoccupied area according to a preset rule, and obtain first pressure data of the chest and abdomen area and second pressure data of the target unoccupied area;

[0008] Perform collaborative filtering processing on the first pressure data through the second pressure data to obtain an adaptive filtering signal;

[0009] Perform band-pass filtering processing on the adaptive filtering signal to obtain a respiratory wave signal;

[0010] Extract peak points from the respiratory wave signal to obtain peak intervals;

[0011] Calculate a respiratory rate value based on the peak intervals.

[0012] In a second aspect, a respiratory rate measurement device is provided, including:

[0013] A chest and abdomen region determination unit, configured to collect pressure map data and determine the chest and abdomen region in the pressure map data;

[0014] A pressure data acquisition unit, configured to determine a target unoccupied region according to a preset rule, and acquire first pressure data of the chest and abdomen region and second pressure data of the target unoccupied region;

[0015] An adaptive filtering acquisition unit, configured to perform collaborative filtering processing on the first pressure data through the second pressure data to obtain an adaptive filtering signal;

[0016] A respiratory wave signal acquisition unit, configured to perform band-pass filtering processing on the adaptive filtering signal to obtain a respiratory wave signal;

[0017] A peak interval acquisition unit, configured to extract peak points from the respiratory wave signal to obtain a peak interval;

[0018] A respiratory rate value acquisition unit, configured to calculate a respiratory rate value based on the peak interval.

[0019] In a third aspect, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the steps of the respiratory rate measurement method as described above are implemented.

[0020] In a fourth aspect, a readable storage medium is provided. The readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the respiratory rate measurement method as described above are implemented.

[0021] For the above-mentioned respiratory rate measurement method, device, computer device, and storage medium, the implementation of the method includes: collecting pressure map data, determining the chest and abdomen region in the pressure map data; determining a target unoccupied region according to a preset rule, and acquiring first pressure data of the chest and abdomen region and second pressure data of the target unoccupied region; performing collaborative filtering processing on the first pressure data through the second pressure data to obtain an adaptive filtering signal; performing band-pass filtering processing on the adaptive filtering signal to obtain a respiratory wave signal; extracting peak points from the respiratory wave signal to obtain a peak interval; calculating a respiratory rate value based on the peak interval. In the embodiments of the present application, by extracting the position where the chest and abdomen are located and using the pressure signal of the unoccupied region to perform collaborative filtering on the pressure signal at the position where the chest and abdomen are located, the signal quality is improved to a certain extent, the accuracy of the respiratory rate is increased, and the respiratory rate measurement can be realized in complex scenarios such as multiple people and multiple people and multiple objects. The application scenarios are wide and the accuracy is higher. Brief Description of the Drawings

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0023] Figure 1 is a schematic flowchart of a respiration rate measurement method in an embodiment of the present invention;

[0024] Figure 2 is a network architecture diagram of a deep learning model in an embodiment of the present invention;

[0025] Figure 3 is a schematic diagram of the chest and abdomen area when the user is in a supine position in an embodiment of the present invention;

[0026] Figure 4 is a schematic diagram of the chest and abdomen area when the user is in a prone position in an embodiment of the present invention;

[0027] Figure 5 is a schematic diagram of the chest and abdomen areas of two users in a lateral lying position respectively in an embodiment of the present invention;

[0028] Figure 6 is a schematic diagram of the chest and abdomen area and the unoccupied area during the user's sleep in an embodiment of the present invention;

[0029] Figure 7 is a change curve graph of the original pressure signal in an embodiment of the present invention;

[0030] Figure 8 is a change curve graph of the pressure signal in the chest and abdomen area in an embodiment of the present invention;

[0031] Figure 9 is a change curve graph of the pressure signal in the unoccupied area in an embodiment of the present invention;

[0032] Figure 10 is a change curve graph of the adaptive filtering signal in an embodiment of the present invention;

[0033] Figure 11 is a schematic change curve diagram for selecting respiration peak points in an embodiment of the present invention;

[0034] Figure 12 is a change curve graph of the measured respiration rate in an embodiment of the present invention;

[0035] Figure 13It is a schematic structural diagram of a breathing rate measurement device in an embodiment of the present invention;

[0036] Figure 14 It is a schematic diagram of a computer device in an embodiment of the present invention. Detailed implementation manners

[0037] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0038] In one embodiment, as Figure 1 shown, a flowchart for implementing a breathing rate measurement method is provided, including the following steps:

[0039] In step S110, pressure map data is collected, and the chest and abdomen regions in the pressure map data are determined;

[0040] In the embodiments of the present application, a pressure detection device may be set on an object for the user to rest, such as a pressure blanket may be set on a smart mattress, a sofa, a bed and other devices on which the user can lie, or a plurality of pressure detection devices arranged in an array may be set. The pressure detection devices may be arranged at equal intervals in an array longitudinally and horizontally, so as to collect the pressure map data.

[0041] It can be understood that the pressure map data may include a pressure image of the area where the human body is located and a pressure image of the unoccupied area. Among them, the pressure image of the area where the human body is located can be obtained by connecting the pressure detection devices that obtain the pressure data, and then the image of the area where the human body is located can be obtained. The unoccupied area may refer to the area where the pressure detection device does not detect pressure data, that is, the area without human contact and without heavy object pressure.

[0042] Among them, the pressure detection device may be a pressure sensor.

[0043] In the embodiments of the present application, the pressure map data may be input into a pre-trained deep learning model for detection to obtain the chest and abdomen position information, such as the abscissa and ordinate of the center point of the chest and abdomen, the width of the chest and abdomen frame, the height of the chest and abdomen frame, etc. Based on the chest and abdomen position information, the area where the chest and abdomen are located can be extracted. At the same time, the pre-trained deep learning model can also output the sleep posture type, such as supine / prone, side lying and others.

[0044] It can be understood that there may be multiple chest and abdomen regions, for example, in the scenario of two people in bed or multiple people in bed, in which case multiple chest and abdomen regions can be extracted to detect the breathing rates of multiple users simultaneously.

[0045] Furthermore, the output result of the pre-trained deep learning model can also determine the number of users. Thus, based on this number of users, it can be determined whether it is consistent with the number of obtained chest and abdomen regions, so as to avoid the problem of detection errors, or detecting other objects as the chest and abdomen regions of a human body, such as an animal.

[0046] In an embodiment of the present application, the network architecture of the deep learning model may specifically be as Figure 2 shown, where the specific operations of the convolutional layer include three steps: Conv2d (convolution), BN (batch normalization), and ReLU (activation). K5 represents a convolutional kernel size of 5*5, K3 represents a convolutional kernel size of 3*3, s2 represents a stride of 2, s1 represents a stride of 1, p1 represents padding type maxpooling, c16 represents 16 convolutional kernels, c32 represents 32 convolutional kernels, c64 represents 64 convolutional kernels, and 32*16*32 represents the output parameter quantity and form of this layer. The fully connected layer corresponds to Fully connectedlayer, upsampling x2 represents upsampling by 2 times, upsampling x4 represents upsampling by 4 times, downsampling x2 represents downsampling by 2 times, and downsampling x4 represents downsampling by 4 times. It includes two operations: summation and ReLU. The basic block is the Basic Block layer in ResNet. The convolutional layer 1*1 is a convolutional layer with a size of 1*1, which is used to adjust the input size and shape and increase non-linearity (it also includes the ReLU operation). The pressure map data is input through the input layer Input layer, with an input parameter of 64*31*1. Then, after being processed by Conv2d (convolution), it is input into the Basic Block layer in ResNet for processing, and then sent to Conv1*1 for convolutional processing. After classification processing through the fully connected layer, multiple features are output, which may specifically include the position information of the chest and abdomen. Among them, the chest and abdomen position information may include the abscissa and ordinate of the center point of the chest and abdomen, the width of the chest and abdomen frame, and the height of the chest and abdomen frame.

[0047] See Figures 3 - 5 , which shows the regions where the chest and abdomen are located in different sleeping postures. In the figure, the regions where the chest and abdomen are located are framed with white boxes. Figure 3 is the region where the chest and abdomen are located when the user is in the supine state. Figure 4 is the region where the chest and abdomen are located when the user is in the prone state. Figure 5 is the region where the chest and abdomen are located when two people are in the side-lying state respectively.

[0048] In the embodiments of the present application, a sample data set can be collected. The sample data set can include pre-collected pressure map sample data. The chest and abdomen regions can be manually marked out. Through the pre-collected pressure map sample data, the deep learning model is iteratively trained until the deep learning model meets the preset convergence conditions. For example, when the number of iterations exceeds a preset number or the loss value is lower than a preset loss threshold, a trained deep learning model can be obtained for subsequent extraction of the chest and abdomen regions.

[0049] It can be understood that during the use of the deep learning model, its model parameters can be continuously optimized to obtain an optimal deep learning model.

[0050] In step S120, determine the target unoccupied area according to a preset rule, and obtain the first pressure data of the chest and abdomen region and the second pressure data of the target unoccupied area.

[0051] In the embodiments of the present application, the area where the human body is located and the target unoccupied area can be determined by obtaining the pressure data detected by each pre-set pressure detection device. Among them, the area where the pressure detection device with pressure data change is located is the area where the human body is located, and the area where the pressure detection device without pressure data change is located is the target unoccupied area. Based on this area where the human body is located, the chest and abdomen regions of the human body can be further determined. Specifically, see Figure 6 , which shows an example of the pressure map of the user on the bed, where the target unoccupied area and the chest and abdomen region of the user are respectively framed with white boxes.

[0052] Further, after determining the chest and abdomen regions and the target unoccupied area, the pressure values collected by each pressure detection device in the chest and abdomen regions at the current moment can be obtained, and then added up and averaged to obtain the first pressure data. Similarly, the second pressure data of the target unoccupied area can also be obtained by calculating the pressure values collected by each pressure detection device in the target unoccupied area and calculating the average value. Among them, during the user's sleep, the first pressure data can be specifically as Figure 8 shown, Figure 8 In the upper figure above, it is the pressure data change curve graph of the chest and abdomen region within 0 - 3000s, and in the lower figure below, it is an enlarged detailed display graph of the pressure change within 0 - 120s in the chest and abdomen region. The second pressure data can be specifically as Figure 9 shown, Figure 9 In the upper figure above, it is the pressure data change curve graph of the unoccupied area within 0 - 3000s, and in the lower figure below, it is an enlarged detailed display graph of the pressure change within 0 - 120s in the unoccupied area. The original pressure data can be as Figure 7 shown, Figure 7The upper middle figure is the pressure change curve of the entire pressure area within 0 - 3000s, and the lower figure is the enlarged detailed display of the pressure change within 0 - 120s.

[0053] It can be understood that the area of the target unoccupied area can be the same as the area of the chest and abdomen area. For example, according to the length and width of the chest and abdomen area, the length and width of the unoccupied area are determined, and then the target unoccupied area that meets the length and width can be extracted from multiple unoccupied areas.

[0054] In step S130, the first pressure data is processed by collaborative filtering using the second pressure data to obtain an adaptive filtering signal;

[0055] In the embodiment of the present application, after obtaining the first pressure data and the second pressure data, the first pressure data can be processed by collaborative filtering using the second pressure data to obtain the adaptive filtering signal, that is, by Figure 9 the second pressure data shown Figure 8 to perform collaborative filtering on the first pressure data shown, thereby obtaining an adaptive filtering signal.

[0056] It should be noted that although there are no people and no heavy objects in the unoccupied area, there will still be relatively small pressure values, mainly due to electromagnetic and other noise interferences in the environment. Therefore, using the pressure data of the unoccupied area to filter the pressure data of the chest and abdomen area can filter out environmental interference noise.

[0057] Among them, the filtering method includes but is not limited to RLS (Recursive Least Squares) and LMS (least mean square). The specific form of the adaptive filtering signal after filtering can be as Figure 10 shown. The upper part shows the change curve of the filtering signal within 0 - 3000s, and the lower part is the detailed magnification display of the filtering signal within 0 - 120s.

[0058] In step S140, the adaptive filtering signal is processed by band - pass filtering to obtain a respiratory wave signal;

[0059] In the embodiment of the present application, after obtaining the adaptive filtering signal, the adaptive filtering can be processed by band - pass filtering using a band - pass filter. The types of the band - pass filter include but are not limited to second - order band - pass filtering, resonant filters, and IIR - type butterworth filters. Its band - pass frequency range can be [0.1, 1.5]Hz, allowing signals within the pass - band frequency range to pass through, while blocking or weakening signals of other frequencies, so as to obtain a respiratory wave signal.

[0060] In step S150, peak points of the respiratory wave signal are extracted to obtain the peak intervals.

[0061] In an embodiment of the present application, after band-pass filtering, a respiratory wave signal can be obtained. By detecting the maximum points and minimum points of the respiratory wave signal, and based on the maximum points and minimum points, the peak points of the respiratory wave signal are obtained. The selected peak points are specifically as Figure 11 shown. In the figure, the positions of the peak points are indicated by hollow circles. Based on the positions of the peak points, the time intervals between adjacent peak points can be obtained, that is, the peak intervals.

[0062] In step S160, based on the peak intervals, the respiratory rate value is calculated.

[0063] In an embodiment of the present application, based on the peak intervals, the stability characteristics of the respiratory wave signal can be calculated, namely the peak amplitude stability, the time interval stability, and the high-frequency energy stability. The high-frequency energy stability can be calculated by the ratio of the first high-frequency energy in the chest and abdomen area to the second high-frequency energy in the unoccupied area. Based on the sum value of the three stabilities, the stability characteristic is obtained, and the respiratory value is calculated according to the following formula:

[0064]

[0065] where BR(i)′ is the respiratory value at the current moment, min represents the minimum operation, BR(i - 1)′ is the respiratory value at the previous moment, BR(i) is the respiratory value before processing at the current moment, and its value is 60 / T, T is the time interval between the current peaks, and β represents the stability characteristic.

[0066] It can be understood that filtering processing, such as Kalman filtering, is performed on the obtained respiratory value to achieve smoothing processing of the respiratory value, and then the real-time output respiratory value, that is, the respiratory rate value, can be obtained, specifically as Figure 12 shown.

[0067] An embodiment of the present application provides a respiratory rate measurement method, including: collecting pressure map data, determining the chest and abdomen area in the pressure map data; determining a target unoccupied area according to a preset rule, obtaining first pressure data of the chest and abdomen area and second pressure data of the target unoccupied area; performing collaborative filtering processing on the first pressure data through the second pressure data to obtain an adaptive filtering signal; performing band-pass filtering processing on the adaptive filtering signal to obtain a respiratory wave signal; extracting peak points from the respiratory wave signal to obtain peak intervals; and calculating a respiratory rate value based on the peak intervals. In the embodiment of the present application, by extracting the position where the chest and abdomen are located and using the pressure signal of the unoccupied area to perform collaborative filtering on the pressure signal at the position where the chest and abdomen are located, the signal quality is improved to a certain extent, the accuracy of the respiratory rate is increased, and the respiratory rate can be measured in complex scenarios such as multiple people and multiple people and multiple objects, with a wide range of adaptable scenarios and higher accuracy.

[0068] In an embodiment of the present application, the determining a target unoccupied area according to a preset rule includes:

[0069] Constructing a pressure map coordinate system, and selecting the coordinate of the center position of a candidate unoccupied area in the pressure map coordinate system;

[0070] Calculating the sum of the pressure values of each point in the candidate unoccupied area and the number of points in the candidate unoccupied area whose pressure values are greater than a first preset threshold;

[0071] When the sum of the pressure values of each point is less than a second preset threshold and the number of points in the candidate unoccupied area whose pressure values are greater than the first preset threshold is less than a third preset threshold, determining the candidate unoccupied area as the target unoccupied area.

[0072] Specifically, a pressure map coordinate system can be constructed according to the pressure map data. For example, taking the vertex at the lower left corner of the pressure matrix as the coordinate origin, the horizontal right direction as the positive direction of the horizontal axis, and the vertical upward direction as the positive direction of the vertical axis, the center position of the candidate unoccupied area can be initialized in the pressure map coordinate system. For example, the center position coordinates of the chest and abdomen are (x0, y0), and the initial candidate unoccupied area center position is where w is the width of the candidate unoccupied area frame. Calculate the sum P of the pressure signal values within the candidate unoccupied area sum and the number Count of the pressure signal values in the candidate unoccupied area that are greater than the preset value th , if P sum is less than the preset value and Count th is less than the preset value, then determine this candidate unoccupied area as the target unoccupied area, otherwise continue the search.

[0073] Among them, the rule for searching the unoccupied area can be from left to right and from y0 to both sides in the above pressure map. If the area satisfies P sumLess than the preset value and Count th If it is less than the preset value, stop the search, and use this area as the above-mentioned unoccupied area. If the above conditions are not met, use the area with the smallest signal sum as the above-mentioned unoccupied area. Specifically, taking a normal sleep segment as an example, the pressure map data collected can be as shown in Figure 6 shown, the selected target unoccupied area frame and the area frame where the chest and abdomen are located are as shown in Figure 6 the white frame shown in

[0074] In an embodiment of the present application, the extracting peak points from the respiratory wave signal to obtain the peak interval includes:

[0075] Detecting the minimum points of the respiratory wave signal. When the minimum value corresponding to the detected minimum point is less than the fourth preset threshold, and the time interval between the minimum point and the previous effective minimum point is within the preset time range, then the minimum point is used as an effective minimum point;

[0076] Detecting the maximum points of the respiratory wave signal. When the maximum value corresponding to the detected maximum point is greater than the fifth preset threshold, and there is an effective minimum point before the maximum point, then the maximum point is used as a peak point.

[0077] Specifically, the minimum points of the respiratory wave signal can be detected. The minimum point refers to the value of the first point on both sides of this point being larger than the value of this point. When the minimum point is detected, it can be determined whether the minimum value corresponding to the minimum point is less than the fourth preset threshold TH1, and whether the time interval between the minimum point and the previous minimum point less than the fourth preset threshold is within the preset time range, such as 2 seconds - 8 seconds. If it meets the conditions, the flag of this minimum point can be set to 1. At the same time, the maximum points of the respiratory wave signal can be detected. The maximum value refers to the value of the first point on both sides of this point being smaller than the value of this point. After the maximum point is detected, if the maximum value corresponding to the maximum point is greater than the fifth preset threshold TH2, and there is an effective minimum point before the maximum point, that is, there is a minimum point with a flag of 1, then the maximum point is used as the wave crest point, and the flag of this minimum point is set to 0.

[0078] Among them, the time interval between this minimum point and the previous minimum point less than the fourth preset threshold refers to the respiratory interval.

[0079] It can be understood that whether there is an effective minimum point before the maximum point means that there is an effective minimum point within a certain time range before the maximum point, and the minimum point is adjacent to the maximum point. The minimum point and the previous effective minimum point can also be adjacent.

[0080] In an embodiment of the present application, the calculating the respiratory rate value based on the peak interval includes:

[0081] Calculate the stability feature of the respiration signal of the target user based on the peak interval;

[0082] Calculate the respiration value at the current moment based on the stability feature;

[0083] Perform filtering processing on the respiration value to obtain the respiration rate value.

[0084] Specifically, based on the peak interval, the stability feature β of the respiration wave signal can be calculated, and the respiration value can be calculated according to the following formula:

[0085]

[0086] Where BR(i)' is the respiration value at the current moment, min is the minimum operation, BR(i - 1)' is the respiration value at the previous moment, BR(i) is the respiration value before processing at the current moment, its value is 60 / T, T is the time interval between the current wave peaks, and β represents the stability feature.

[0087] It can be understood that filtering processing, such as Kalman filtering, is performed on the obtained respiration value to achieve smoothing processing of the respiration value, and the real-time output respiration value, that is, the respiration rate value, can be obtained.

[0088] Further, calculating the stability feature of the respiration signal of the target user includes:

[0089] Calculate the peak amplitude stability, time interval stability, and high-frequency energy stability respectively;

[0090] Calculate the sum value of the peak amplitude stability, time interval stability, and high-frequency energy stability as the stability feature of the respiration signal.

[0091] Specifically, the signal stability feature includes peak amplitude stability, time interval stability, and high-frequency energy stability. Through the peak interval, the peak amplitude stability β1, time interval stability β2, and high-frequency energy stability β3 are calculated respectively, and then the signal stability feature is obtained based on the sum value of the three, that is, β = β = β1 + β2 + β3.

[0092] Among them, the peak amplitude stability can be obtained through the following method:

[0093] Perform median filtering on the peak amplitudes within the first preset time window, and calculate the first mean value of the peak amplitudes after median filtering;

[0094] Calculate the first ratio between the peak amplitude at the current moment and the first mean value;

[0095] Calculate the crest amplitude stability based on the first ratio.

[0096] Specifically, median filtering can be performed on the crest amplitudes within the first preset time window, for example, from the current time to 60 seconds before it. The length of this time window can be 3. Then, calculate the first mean value of each crest amplitude after median filtering, and calculate the first ratio α1 between the crest amplitude at the current moment and this first mean value. Based on this first ratio α1, calculate the crest amplitude stability β1 according to the following formula:

[0097]

[0098] Among them, the time interval stability can be obtained in the following manner:

[0099] Perform median filtering on the time intervals within the second preset time window, and calculate the second mean value of the time intervals after median filtering;

[0100] Calculate the second ratio between the time interval at the current moment and the second mean value;

[0101] Calculate the time interval stability based on the second ratio.

[0102] Specifically, median filtering can be performed on the time intervals within the second preset time window, for example, from the current time to 60 seconds before it, that is, the time intervals between adjacent crest points. Then, calculate the second mean value of all the time intervals after median filtering. Next, calculate the second ratio α2 between the time interval at the current moment and this second mean value, and calculate the time interval stability β3 according to the following formula based on this second ratio α2:

[0103]

[0104] Among them, the high-frequency energy stability is obtained in the following manner:

[0105] Calculate the first high-frequency energy of the first pressure data and the second high-frequency energy of the second pressure data;

[0106] Calculate the high-frequency energy stability based on the third ratio between the first high-frequency energy and the second high-frequency energy.

[0107] Specifically, calculate the corresponding first high-frequency energy and second high-frequency energy for the first pressure data in the chest and abdomen area and the second pressure data in the unoccupied area respectively. Then, calculate the third ratio α3 between the first high-frequency energy and this second high-frequency energy, and calculate the high-frequency energy stability β3 according to the following formula based on this third ratio α3:

[0108]

[0109] Among them, calculating the first high-frequency energy of the first pressure data and the second high-frequency energy of the second pressure data includes:

[0110] Performing high-pass filtering on the first pressure data and the second pressure data respectively;

[0111] Calculating the sum of the absolute values of the first pressure data after high-pass filtering within a third preset time window as the first high-frequency energy;

[0112] Calculating the sum of the absolute values of the second pressure data after high-pass filtering within a third preset time window as the second high-frequency energy.

[0113] Specifically, for the first pressure data and the second pressure data, high-pass filtering can be performed through a high-pass filter. The types of this high-pass filter can specifically include a first-order RC high-pass filter, a multifunctional second-order general filter, etc. The first pressure data and the second pressure data can be smoothed, and then the sum of the absolute values of the first pressure data after high-pass filtering is calculated and used as the first high-frequency energy, while the sum of the absolute values of the second pressure data after high-pass filtering is used as the second high-frequency energy.

[0114] In the embodiments of the present application, by extracting the positions of the chest and abdomen and using the pressure signals in the unoccupied area to perform collaborative filtering on the pressure signals at the positions of the chest and abdomen, the signal quality is improved to a certain extent, the accuracy of the respiratory rate is increased, and the respiratory rate can be measured in complex scenarios such as multiple people and multiple people and multiple objects. It has a wide range of applicable scenarios and higher accuracy.

[0115] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0116] In one embodiment, a respiratory rate measurement device is provided, and this respiratory rate measurement device corresponds one-to-one to the respiratory rate measurement method in the above embodiment. As Figure 13 shown, this respiratory rate measurement device includes a chest and abdomen area determination unit 10, a pressure data acquisition unit 20, an adaptive filtering acquisition unit 30, a respiratory wave signal acquisition unit 40, a peak interval acquisition unit 50, and a respiratory rate value acquisition unit 60. The detailed description of each functional module is as follows:

[0117] The chest and abdomen area determination unit 10 is used to collect pressure map data and determine the chest and abdomen area in the pressure map data;

[0118] A pressure data acquisition unit 20, configured to determine a target unoccupied area according to a preset rule, and acquire first pressure data of the chest and abdomen area and second pressure data of the target unoccupied area;

[0119] An adaptive filtering acquisition unit 30, configured to perform collaborative filtering processing on the first pressure data through the second pressure data to obtain an adaptive filtering signal;

[0120] A respiratory wave signal acquisition unit 40, configured to perform band-pass filtering processing on the adaptive filtering signal to obtain a respiratory wave signal;

[0121] A peak interval acquisition unit 50, configured to extract peak points from the respiratory wave signal to obtain a peak interval;

[0122] A respiratory rate value acquisition unit 60, configured to calculate a respiratory rate value based on the peak interval.

[0123] In an embodiment of the present application, the pressure data acquisition unit 20 is further configured to:

[0124] Construct a pressure map coordinate system, and select the coordinates of the center position of the candidate unoccupied area in the pressure map coordinate system;

[0125] Calculate the sum of the pressure values of each point in the candidate unoccupied area and the number of points in the candidate unoccupied area whose pressure values are greater than a first preset threshold;

[0126] When the sum of the pressure values of each point is less than a second preset threshold and the number of points in the candidate unoccupied area whose pressure values are greater than the first preset threshold is less than a third preset threshold, the candidate unoccupied area is used as the unoccupied area.

[0127] In an embodiment of the present application, the peak interval acquisition unit 50 is further configured to:

[0128] Detect minimum value points of the respiratory wave signal. When the minimum value corresponding to the detected minimum value point is less than a fourth preset threshold and the time interval between the minimum value point and the previous effective minimum value point is within a preset time range, the minimum value point is used as an effective minimum value point;

[0129] Detect maximum value points of the respiratory wave signal. When the maximum value corresponding to the detected maximum value point is greater than a fifth preset threshold and there is an effective minimum value point before the maximum value point, the maximum value point is used as a peak point.

[0130] In an embodiment of the present application, the respiratory rate value acquisition unit 60 is further configured to:

[0131] Calculate a respiratory signal stability feature of the target user based on the peak interval;

[0132] Calculate the breathing value at the current moment based on the stability feature;

[0133] Perform filtering processing on the breathing value to obtain the breathing rate value.

[0134] In an embodiment of the present application, the breathing rate value acquisition unit 60 is further configured to:

[0135] Calculate the peak amplitude stability, time interval stability, and high-frequency energy stability respectively;

[0136] Calculate the sum value of the peak amplitude stability, time interval stability, and high-frequency energy stability as the breathing signal stability feature.

[0137] In an embodiment of the present application, the breathing rate value acquisition unit 60 is further configured to:

[0138] Perform median filtering on the peak amplitude within the first preset time window, and calculate the first mean value of the peak amplitude after median filtering;

[0139] Calculate the first ratio between the peak amplitude at the current moment and the first mean value;

[0140] Calculate the peak amplitude stability based on the first ratio.

[0141] In an embodiment of the present application, the breathing rate value acquisition unit 60 is further configured to:

[0142] Perform median filtering on the time interval within the second preset time window, and calculate the second mean value of the time interval after median filtering;

[0143] Calculate the second ratio between the time interval at the current moment and the second mean value;

[0144] Calculate the time interval stability based on the second ratio.

[0145] In an embodiment of the present application, the breathing rate value acquisition unit 60 is further configured to:

[0146] Calculate the first high-frequency energy of the first pressure data and the second high-frequency energy of the second pressure data;

[0147] Calculate the high-frequency energy stability based on the third ratio between the first high-frequency energy and the second high-frequency energy.

[0148] In an embodiment of the present application, the breathing rate value acquisition unit 60 is further configured to:

[0149] Perform high-pass filtering on the first pressure data and the second pressure data respectively;

[0150] Calculate the sum of the absolute values of the first pressure data after high-pass filtering within a third preset time window as the first high-frequency energy;

[0151] Calculate the sum of the absolute values of the second pressure data after high-pass filtering within a third preset time window as the second high-frequency energy.

[0152] In the embodiments of the present application, by extracting the positions of the chest and abdomen and using the pressure signals in the unoccupied area to perform collaborative filtering on the pressure signals at the positions of the chest and abdomen, the signal quality is improved to a certain extent, the accuracy of the respiratory rate is increased, and the respiratory rate can be measured in complex scenarios such as multiple people and multiple people and multiple objects. The applicable scenarios are wide and the accuracy is higher.

[0153] For the specific limitations of the respiratory rate measurement device, reference can be made to the limitations on the respiratory rate measurement method in the above text, which will not be elaborated here. Each module in the above respiratory rate measurement device can be implemented in whole or in part by software, hardware, and their combinations. The above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0154] In one embodiment, a computer device is provided. The computer device can be a terminal device, and its internal structure diagram can be as Figure 14 shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium. The readable storage medium stores computer-readable instructions. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instructions are executed by the processor, a respiratory rate measurement method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0155] In the embodiments of the present application, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the steps of the respiratory rate measurement method as described above are implemented.

[0156] In the embodiments of the application, a readable storage medium is provided. The readable storage medium stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps of the respiratory rate measurement method as described above are implemented.

[0157] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through computer-readable instructions. The computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0158] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0159] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for measuring respiratory rate, characterized in that: The method comprises: Collecting pressure map data and determining the chest and abdomen regions in the pressure map data; Determine the target unmanned area according to a preset rule, and obtain the first pressure data of the chest and abdomen area and the second pressure data of the target unmanned area; Performing collaborative filtering processing on the first pressure data through the second pressure data to obtain an adaptive filtering signal; Performing bandpass filtering on the adaptive filtering signal to obtain a respiratory wave signal; Extracting peak points of the respiratory wave signal to obtain peak intervals; Based on the peak interval, calculating a breathing signal stability feature of the target user; Based on the stability characteristic, calculating the current breathing value; Performing filtering on the respiration value to obtain a respiration rate value; The respiration value is calculated by the following formula: in is the breathing value at the current moment, min is the minimum operation, is the breathing value at the previous moment, is the breathing value before processing at the current moment, and its value is 60 / T, where T is the time interval between the current peaks. , , is the peak amplitude stability, is the time interval stability, For high frequency energy stability.

2. The respiratory rate measurement method according to claim 1, characterized in that: The step of determining the target unmanned area according to a preset rule includes: Constructing a pressure map coordinate system, and selecting the coordinates of the center position of the candidate unmanned area in the pressure map coordinate system; Calculate the sum of the pressure values ​​of each point in the candidate unmanned area and the number of pressure values ​​of each point in the candidate unmanned area that are greater than a first preset threshold; When the sum of the pressure values ​​at the points is less than the second preset threshold, and the number of the pressure values ​​at the points greater than the first preset threshold is less than the third preset threshold, the candidate unmanned area is taken as the target unmanned area.

3. The respiratory rate measurement method according to claim 1, characterized in that: The step of extracting the peak point of the respiratory wave signal to obtain the peak interval includes: Performing a minimum point detection on the respiratory wave signal, when the minimum value corresponding to the detected minimum point is less than a fourth preset threshold value, and the time interval between the minimum point and the previous valid minimum point is within a preset time range, the minimum point is taken as a valid minimum point; A maximum point detection is performed on the respiratory wave signal. When a maximum value corresponding to the detected maximum point is greater than a fifth preset threshold and there is a valid minimum point before the maximum point, the maximum point is taken as a peak point.

4. The respiratory rate measurement method according to claim 1, characterized in that: Calculating the breathing signal stability feature of the target user, including: Calculate the peak amplitude stability, time interval stability and high-frequency energy stability respectively; The sum of the peak amplitude stability, the time interval stability and the high-frequency energy stability is calculated as the stability feature of the respiratory signal.

5. The respiratory rate measurement method according to claim 4, characterized in that: The peak amplitude stability is obtained by: Performing median filtering on the peak amplitude within the first preset time window, and calculating a first mean value of the peak amplitude after the median filtering; Calculate a first ratio between the peak amplitude at the current moment and the first mean value; The peak amplitude stability is calculated based on the first ratio.

6. The respiratory rate measurement method according to claim 4, characterized in that: The time interval stability is obtained by: Performing median filtering on the time intervals within the second preset time window, and calculating a second mean value of the time intervals after the median filtering; Calculate a second ratio between the current time interval and the second mean; Based on the second ratio, the time interval stability is calculated.

7. The respiratory rate measurement method according to claim 4, characterized in that: The high frequency energy stability is obtained by: calculating a first high-frequency energy of the first pressure data and a second high-frequency energy of the second pressure data; Based on a third ratio between the first high-frequency energy and the second high-frequency energy, high-frequency energy stability is calculated.

8. The respiratory rate measurement method according to claim 7, characterized in that: The calculating the first high-frequency energy of the first pressure data and the second high-frequency energy of the second pressure data includes: performing high-pass filtering on the first pressure data and the second pressure data respectively; calculating the sum of absolute values ​​of the first pressure data after high-pass filtering within a third preset time window as the first high-frequency energy; The sum of the absolute values ​​of the second pressure data after high-pass filtering within the third preset time window is calculated as the second high-frequency energy.

9. A respiratory rate measuring device, characterized in that: The device for implementing the respiratory rate measurement method according to any one of claims 1 to 8 comprises: A chest and abdomen region determination unit, used to collect pressure map data and determine the chest and abdomen region in the pressure map data; A pressure data acquisition unit, configured to determine a target unmanned area according to a preset rule, and acquire first pressure data of the chest and abdomen area and second pressure data of the target unmanned area; an adaptive filtering acquisition unit, configured to perform collaborative filtering processing on the first pressure data through the second pressure data to obtain an adaptive filtering signal; A respiratory wave signal acquisition unit, used for performing bandpass filtering on the adaptive filtering signal to obtain a respiratory wave signal; A peak interval acquisition unit, used for extracting peak points of the respiratory wave signal to obtain a peak interval; The breathing rate value acquisition unit is used to calculate the stability characteristics of the target user's breathing signal based on the peak interval; calculate the breathing value at the current moment based on the stability characteristics; and filter the breathing value to obtain the breathing rate value.

Images