Human Feature Data Acquisition Methods and Devices

Abstract

This application discloses a method and apparatus for acquiring human feature data. The method includes: using an ultra-wideband (UWB) radar array, the acquisition device performs real-time identification of a target human body within a signal-detectable area, and determines a three-dimensional upper body model for the identified target human body; based on the three-dimensional upper body model, the chest and abdomen region of the target human body is determined, and the chest and abdomen region is detected; based on the set of data points obtained from the detection, a contour model of the chest and abdomen region and monitoring control points in the contour model are determined; the monitoring control points include respiratory monitoring control points and / or heart rate monitoring control points; using the UWB radar array, the chest and abdomen region is detected in real time, and based on the coordinate changes of the detected monitoring control points, the respiratory rate and / or heart rate of the target human body are determined. Using this application can improve the comfort of the user being tested during human feature data acquisition.

Description

Technical Field

[0001] This invention relates to human physiological characteristic monitoring technology, and in particular to a method and device for acquiring human characteristic data. Background Technology

[0002] Currently, the physiological characteristics of human heartbeat and respiration are usually monitored using instruments such as cardiac monitors and electrocardiogram monitors, as well as equipment such as heart rate monitors, breathing straps, and stethoscopes.

[0003] In the process of developing this invention, the inventors discovered that existing physiological characteristic monitoring methods suffer from poor user experience. This is because these methods utilize contact-based examination equipment, meaning the equipment needs to come into contact with the human skin. For patients requiring prolonged monitoring, or individuals with unhealed skin lesions or skin allergies, prolonged contact with the testing equipment can be inconvenient or harmful. Therefore, existing physiological characteristic monitoring methods, relying on direct contact between the equipment and the subject, cause discomfort or inconvenience to the subject during the collection of human characteristic data. Summary of the Invention

[0004] In view of this, the main objective of the present invention is to provide a method and apparatus for collecting human body feature data, which can improve the comfort of the user being tested during the collection of human body feature data.

[0005] To achieve the above objectives, the technical solution proposed in this embodiment of the invention is as follows:

[0006] A method for collecting human feature data, comprising:

[0007] The acquisition equipment uses an ultra-wideband (UWB) radar array to identify target human bodies in real time within the signal detection area, and to determine a three-dimensional model of the upper body for the identified target human body.

[0008] Based on the upper body 3D model, the chest and abdomen region of the target human body is determined, and the chest and abdomen region is detected; based on the set of data points obtained from the detection, the contour model of the chest and abdomen region and the monitoring control points in the contour model are determined; the monitoring control points include respiratory monitoring control points and / or heart rate monitoring control points;

[0009] Using a UWB radar array, the chest and abdominal region is detected in real time, and the respiratory rate and / or heart rate of the target human body are determined based on the detected coordinate changes of the monitoring and control points.

[0010] This invention also proposes a human feature data acquisition device, comprising:

[0011] The target recognition unit is used to identify target human bodies in real time within the signal detectable area using an ultra-wideband (UWB) radar array, and to determine a three-dimensional upper body model for the identified target human body.

[0012] The detection preparation unit is used to determine the chest and abdomen region of the target human body based on the upper body three-dimensional model, and to detect the chest and abdomen region; based on the set of data points obtained from the detection, it determines the contour model of the chest and abdomen region and the monitoring control points in the contour model; the monitoring control points include respiratory monitoring control points and / or heart rate monitoring control points;

[0013] The data acquisition unit is used to perform real-time detection of the chest and abdominal region using a UWB radar array, and to determine the respiratory rate and / or heart rate of the target human body based on the detected coordinate changes of the monitoring and control points.

[0014] This invention also proposes a human feature data acquisition device, including a processor and a memory;

[0015] The memory stores an application program that can be executed by the processor, which enables the processor to perform the human feature data acquisition method described above.

[0016] This invention also proposes a computer-readable storage medium storing computer-readable instructions for performing the human feature data acquisition described above.

[0017] This invention also proposes a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of human feature data acquisition as described above.

[0018] In summary, the human feature data acquisition scheme proposed in this invention utilizes ultra-wideband (UWB) technology to identify the target object and determine a three-dimensional upper body model for the identified target human body. Then, based on this upper body model, a contour model of the chest and abdomen region and monitoring control points are determined for real-time detection of the target human body's physiological characteristics. Finally, a UWB radar array is used to detect the chest and abdomen region in real time, and the respiratory rate and / or heart rate of the target human body are determined based on the coordinate changes of the detected monitoring control points. Thus, the acquisition device accurately identifies and locates the chest and abdomen region of the target human body using UWB signals, eliminating the need for direct contact between the detection device and the target object during human feature data acquisition, thereby improving the comfort of the user being tested. Attached Figure Description

[0019] Figure 1This is a schematic diagram of the method flow according to an embodiment of the present invention;

[0020] Figure 2 This is a schematic diagram illustrating the use of a UWB radar array to determine key detection areas in an embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram of the upper body three-dimensional model of the target human body in an embodiment of the present invention;

[0022] Figure 4 This is a schematic diagram of the contour model of the chest and abdomen region obtained by bending and smoothing thin plate splines in an embodiment of the present invention.

[0023] Figure 5 A diagram illustrating the rise and fall of the chest and abdomen during human respiration.

[0024] Figure 6 This is a schematic diagram of some steps in the deviation compensation process according to an embodiment of the present invention;

[0025] Figure 7 This is a schematic diagram illustrating the process of filling in missing data according to an embodiment of the present invention;

[0026] Figure 8 This is a schematic diagram illustrating the application of an embodiment of the present invention in scenario one;

[0027] Figure 9 This is a schematic diagram illustrating the application of the present invention in scenario two.

[0028] Figure 10 This is a schematic diagram illustrating the application of the present invention in scenario three.

[0029] Figure 11 This is a schematic diagram of the device structure according to an embodiment of the present invention. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0031] Figure 1 This is a schematic diagram of the human feature data acquisition method according to an embodiment of the present invention, such as... Figure 1 As shown, this embodiment mainly includes:

[0032] Step 101: The acquisition device uses an ultra-wideband (UWB) radar array to identify target human bodies in real time within the signal detection area, and determines a three-dimensional upper body model for the identified target human body.

[0033] In this step, the acquisition device is used to identify nearby targets (i.e., the target human body) in real time and determine a 3D model of the upper body. This model is then used in subsequent steps to determine the contour model of the chest and abdomen region for collecting physiological feature data of the target human body, as well as the corresponding monitoring and control points. Thus, by introducing UWB technology for human body recognition and detection, on the one hand, the human feature acquisition device does not need to contact the human body when collecting human feature data, thereby improving user comfort. Furthermore, since the target human body is not limited to a fixed space due to the device's usage scenario, the flexibility of data acquisition is increased, effectively expanding the application scenarios. On the other hand, the penetrability of UWB signals through non-metallic media (clothing, jewelry, etc.) ensures accurate acquisition of human features.

[0034] Specifically, in step 101, a UWB radar array will be used to send an ultra-wideband linear frequency modulated continuous wave signal to the surrounding environment. The signal will be reflected by nearby human bodies or objects to form echoes. Based on these echoes, the acquisition device will use radar imaging technology combined with a human body recognition algorithm to identify nearby human information, thereby obtaining the location and distance information between the device and the target human body. The specific methods for identifying the target human body using radar imaging technology are well-known to those skilled in the art and will not be elaborated upon here.

[0035] The data acquisition device may be a wearable device with an integrated UWB radar array, such as a smartwatch, but is not limited to this.

[0036] In one implementation, the following method can be used to determine a three-dimensional upper body model for the identified target human body:

[0037] Step a1: Determine the current key detection area based on the location of the target human body.

[0038] Specifically, in this step, the location of the identified target human body will be used as the key detection direction of the UWB radar array, and the area where the target human body is located in that direction will be used as the current key detection area (e.g., Figure 2 As shown in the figure, compared with the indiscriminate wide-range detection used by traditional UWB detection methods, the method of identifying the target human body and adjusting the focus of detection can perform detailed detection only on the target human body, which greatly reduces the detection range. This can reduce UWB power consumption, reduce unnecessary computation, and improve detection accuracy and efficiency.

[0039] Step a2: Within the key detection area, a signal is sent to the target human body using a UWB radar array to determine the three-dimensional model of the upper body of the target human body; the three-dimensional model of the upper body includes the skeletal points of each part of the upper limbs and the template data of the human upper limbs.

[0040] Preferably, in order to improve the accuracy of creating a 3D model of the upper body of the target human body, in one embodiment, the 3D model of the upper body of the target human body can be determined by the following method:

[0041] Step a21: Within the key detection area, a UWB radar array is used to send a UWB linear frequency modulated continuous wave signal to the human body.

[0042] Step a22: Based on the echo reflected by the target human body to the UWB linear frequency modulated continuous wave signal, use the mean value method to remove fixed clutter in each distance cell of the echo receiving matrix.

[0043] Here, it is necessary to remove the fixed clutter in each range cell of the echo reception matrix in order to accurately locate the target human body in the range cell in subsequent steps.

[0044] Step a23: Based on the range cell with fixed clutter removed, use constant false alarm rate (CFAR) detection technology to determine the range cell where the target human body is located, and generate three-dimensional range and azimuth direction reception matrix data based on the three-dimensional array echo data of the determined range cell.

[0045] Here, existing CFAR detection technology will be used to detect each range cell after removing fixed clutter, obtaining the range cell where the target human body is located. Then, based on the obtained range cell where the target human body is located, the azimuth direction three-dimensional array echo data of the range cell where the target human body is located will be acquired. After sampling and quantization processing of the three-dimensional array echo data, three-dimensional range-azimuth direction reception matrix data will be obtained. In subsequent steps, based on this three-dimensional range-azimuth direction reception matrix data, a three-dimensional model of the upper body of the target human body will be obtained. The specific methods for determining the three-dimensional array echo data and generating the three-dimensional range and azimuth direction reception matrix data in this step are known to those skilled in the art and will not be described in detail here.

[0046] Step a24: Based on the three-dimensional distance and orientation receiving matrix data, perform data preprocessing to separate different parts of the target human body to obtain the skeletal points of each part of the upper limb of the target human body and the template data of the human upper limb.

[0047] Specifically, in one implementation, such as Figure 3As shown, based on the three-dimensional distance-azimuth receiving matrix data, a random forest algorithm, a preset standard human body template for data preprocessing, and the skeletal information within the standard human body template can be used to separate and preprocess different parts of the target human body to obtain the upper body three-dimensional model. This model serves as the basic three-dimensional data model for real-time detection of the target human body. To improve the accuracy of monitoring and control point determination, the skeletal information within the standard human body template will include pre-defined identification information for respiratory detection areas and / or heartbeat detection areas.

[0048] Step 102: Based on the upper body 3D model, determine the chest and abdomen region of the target human body and detect the chest and abdomen region; based on the set of data points obtained from the detection, determine the contour model of the chest and abdomen region and the monitoring control points in the contour model; the monitoring control points include respiratory monitoring control points and / or heart rate monitoring control points.

[0049] This step is used to determine the contour model of the chest and abdomen region for real-time detection of the target human body, as well as the monitoring control points therein.

[0050] Preferably, in one embodiment, such as Figure 4 As shown, the contour model of the chest and abdomen region can be determined using the following method:

[0051] The data points in the data point set are subjected to thin plate spline (TPS) bending smoothing processing to obtain the contour model of the chest and abdomen region.

[0052] In one implementation, the monitoring and control points can be determined using the following method:

[0053] Based on the respiratory detection area and / or heartbeat detection area in the preset standard human body template, the respiratory monitoring control point and / or heartbeat monitoring control point are determined.

[0054] Preferably, in order to improve the accuracy of data acquisition, one or more data points with the greatest change in distance from the device can be selected within the above-mentioned respiratory detection area or heartbeat detection area based on the set of data points, as the corresponding respiratory or heartbeat monitoring control points.

[0055] Step 103: Using a UWB radar array, the chest and abdominal region is detected in real time, and the respiratory rate and / or heart rate of the target human body are determined based on the detected coordinate changes of the monitoring and control points.

[0056] In this step, the chest and abdominal regions of the target human body determined in step 102 are detected in real time, and the detection results are used to analyze the coordinate changes of the monitoring control points determined in step 102 to obtain physiological characteristic data of the target human body.

[0057] In practical applications, to improve the accuracy of human feature data collection, it is possible to distinguish between the static and dynamic states of the human body for data collection. The specific methods are as follows:

[0058] In one implementation, when the target human body is in a static state, the respiratory rate and / or heart rate data of the target human body can be determined using the following method:

[0059] Based on the coordinate data of the monitoring and control point detected within a specified time range, the reciprocating movement frequency of the monitoring and control point within the time range is determined, and the corresponding respiratory rate and / or heart rate is obtained.

[0060] Here, considering the rise and fall of the chest and abdomen during respiration, the corresponding monitoring control point will shift. Normally, during respiration, the diaphragm relaxes and contracts, causing the diaphragm's center to move forward and backward, resulting in a radial displacement of the thoracic cavity, producing a fluctuation of 0–3 cm. The beating of the heart causes a fluctuation of 1.5–3.5 mm across the surface of the thoracic cavity in the cardiac region. Based on this, the corresponding respiratory and / or heart rate can be obtained based on the reciprocating frequency of the monitoring control point. For example... Figure 5 As shown in the figure, the monitoring control point P has moved to point P'. The lung fluctuation variable can then be calculated using the formula, which represents the distance d' from P to P'. After sampling and calculation over a period of time, the fluctuation frequency of the human chest, abdomen, and heart can be obtained, namely, the respiratory rate and heart rate.

[0061] The specific time range can be set by those skilled in the art based on the accuracy requirements of the actual measurement, and will not be elaborated here.

[0062] In one embodiment, when the target human body is in motion, preferably, in order to improve the accuracy of human characteristic data collection in this state, the following method can be used to determine the respiratory rate and / or heart rate data of the target human body:

[0063] Step x1: For each coordinate data of the monitoring and control point detected within a specified time range, based on the three-dimensional rotation data monitored in real time by the motion sensor of the acquisition device, the coordinate data is subjected to deviation compensation processing to eliminate the positional deviation caused by the relative motion between the target human body and the acquisition device; the three-dimensional rotation data includes the three-dimensional angle and three-dimensional displacement of the acquisition device relative to the target human body.

[0064] It's important to note that real-time monitoring of respiratory and heart rates is more challenging when the human body and the data acquisition device are in relative motion compared to a relatively stationary state. While step 101 above allows for real-time adjustment of the key detection area based on the current human body identification status to offset some of the angle and position changes caused by relative motion, adjusting the detection direction alone cannot completely compensate for deviations caused by significant relative motion. To eliminate this deviation and achieve relatively accurate dynamic monitoring, the method incorporates motion sensor data from the device to overcome the biases caused by relative displacement and angle changes. Therefore, compared to existing UWB devices that can only be fixed in one location for detection, the above method, by introducing motion sensors for deviation compensation, can significantly improve the applicability of devices collecting respiratory and / or heart rate data.

[0065] In practical applications, the deviation compensation process in step x1 can be implemented using existing methods. For example... Figure 6 As shown, assuming v1(a1, b1, c1) and v1(a1, b1, c1) represent the rotation axes of the acquisition device under human motion, and θ represents the rotation angle of the acquisition device under human motion, the deviation compensation processing can be implemented by the following steps through matrix rotation transformation operation.

[0066] 1. Displace one end v1 (a1, b1, c1) of the rotation axis along the x, y, and z axes by -x, -y, and -z respectively to the origin;

[0067] 2. Rotate the axis of rotation by -α degrees along the x-axis to the XOZ plane;

[0068] 3. Rotate the axis of rotation by β degrees along the y-axis to the z-axis;

[0069] 4. Rotate the axis of rotation θ degrees along the y-axis;

[0070] 5. Perform the reverse process of step 3, that is, rotate the axis of rotation along the y-axis by -β degrees;

[0071] 6. Perform the reverse process of step 2, that is, rotate the axis of rotation α degrees along the x-axis;

[0072] 7. Perform the reverse process of step 1, that is, the end v1 (a1, b1, c1) of the rotation axis located at the origin is displaced x, y, z from the origin along the x, y, z axes respectively.

[0073] The above processing can be represented by the following formula:

[0074]

[0075] (u, v, w) = (a2, b2, c2) - (a1, b1, c1), and is a unit vector, where a, b, and c represent (a1, b1, c1) respectively.

[0076] Where T(-x,-y,-z) represents the processing of step 1;

[0077] Rx(-α) represents the processing step 2;

[0078] Ry(β) represents the processing performed in step 3;

[0079] Rz(θ) represents the processing performed in step 4;

[0080] Ry(-β) indicates that step 5 is performed;

[0081] Rx(α) represents the processing step 6;

[0082] T(x,y,z) represents the processing performed in step 7.

[0083] Step x2: Based on the coordinate data after deviation compensation processing within the specified time range, determine the reciprocating movement frequency of the monitoring and control point within the specified time range, and obtain the corresponding respiratory rate and / or heart rate.

[0084] Here, as mentioned above, the displacement deviation of the monitoring control point relative to the acquisition device caused by the movement of the target human body is effectively eliminated through deviation compensation processing. Therefore, the reciprocating movement of the monitoring control point caused by the rise and fall of the chest and abdomen can be effectively determined, thereby improving the accuracy of the reciprocating movement frequency of the monitoring control point obtained under movement conditions within a certain time range.

[0085] Considering that in practical applications, when the target human body is in certain postures, the chest and abdomen may be partially obscured, leading to incomplete chest detection and loss of feature points. In such cases, the incomplete detection data will result in inaccurate feature data acquisition. To address this problem, the following method can be preferably adopted:

[0086] In one implementation, during real-time detection of the chest and abdomen region, an integrity check is performed on the set of data points corresponding to the detected chest and abdomen region. When a data point is missing, an alarm is triggered. When the number of alarms reaches a preset threshold within a preset time range, a matching model is searched from a preset model library based on the set of data points detected within the current time range. The searched model is then used to complete and fill in the set of data points within the time range to obtain the corresponding complete set of data points.

[0087] like Figure 7As shown, if there is a missing detection data, the system will extract the data from the corresponding time period in which the data is missing, search for a similar model in the model library, and fill in the missing data to obtain a relatively accurate respiratory and heart rate.

[0088] In one implementation, the model library can be constructed by training models for different states based on feature data collected from users in different states during real-time detection.

[0089] The aforementioned threshold is used to limit the timing of data point completion and padding. A smaller value makes the system more sensitive to triggering these processes, thus better ensuring the integrity of the data point set used for feature acquisition. However, this may also lead to excessive computational overhead due to overly frequent completion and padding. Specifically, those skilled in the art can set an appropriate threshold based on the accuracy requirements of feature data acquisition in practical applications and by balancing computational costs.

[0090] The time range is used to limit the duration for counting alarms. The shorter the time range, the more timely the completion and padding will be, which can better ensure the integrity of the data point set used for feature collection. However, it may also lead to excessive computational overhead. Specifically, those skilled in the art can set an appropriate time range based on the accuracy requirements of feature data collection in actual applications and weigh the computational overhead.

[0091] In one embodiment, in step 103, the respiratory rate and / or heart rate of the identified target human body can be further judged for abnormality. If abnormal feature data is detected, the user being tested can be alerted by an alarm.

[0092] As can be seen from the above technical solutions, the above method embodiments utilize UWB signals to accurately identify and locate the chest and abdominal regions of the target human body. This eliminates the need for direct contact between the detection device and the target object when collecting human feature data, thereby improving the comfort of the user during data collection. Based on these characteristics, the above method embodiments can be applied to daily human body monitoring, as well as the monitoring of the elderly, children, and patients (especially those with sleep apnea disorders). It can also generate health reports based on monitored respiratory and heart rate, helping users to promptly detect abnormalities. Furthermore, the above method embodiments can also be applied to the monitoring of pets' vital signs, focusing on their health and safety. Simultaneously, the above method embodiments also have excellent applications in virtual scenarios such as the metaverse, where they can be used in virtual avatars (such as AVATAR avatars) during video calls to reflect changes in the bodies of both parties, increasing fun and realism.

[0093] The following examples illustrate the specific applications of this application:

[0094] Scenario 1: Users monitor feature data using wearable devices (such as smartwatches) while at rest.

[0095] like Figure 8 As shown, when the user is resting, the wearable device does not need to be worn. The wearable device can be placed nearby. The wearable device uses a UWB radar array to penetrate bedding and clothing and uses the above-mentioned feature data collection method to detect the user's breathing rate and heart rate during sleep. The data is recorded and a report is generated for the user to view. If there is any abnormal state during the monitoring process (such as sleep apnea), an alarm will be issued in time to remind the user or guardian to deal with it in time.

[0096] Scenario 2: Users monitor feature data using wearable devices (such as smartwatches and wristbands) while in motion.

[0097] like Figure 9 As shown, when a user is in motion, monitoring the user's respiratory rate and heart rate is necessary. Compared to a static state, the relative motion between the wearable device and the user needs to be considered. Relying on the motion sensors of the wearable device and the model data of the human upper limb body, the respiratory and heart rates can be calculated by using matrix rotation and movement algorithms and UWB array detection angle adjustment.

[0098] In cases where the UWB array is blocked by the body and cannot detect the real data due to excessive swing amplitude, inertial inference can be made by relying on the data detected before and after and the accumulated data model to compensate for the data lost due to occlusion in a short period of time.

[0099] Scenario 3: Application of human feature data in the metaverse and virtual reality scenarios.

[0100] like Figure 10 As shown, the above-described method embodiments can be applied to the metaverse and virtual reality. Because virtual humans do not reflect the vital signs of real people, they lack a sense of realism in virtual scenarios such as social interactions or games. Users cannot perceive their own or others' physical changes, such as rapid breathing while running or an accelerated heartbeat when nervous or excited. Using the above-described method embodiments, the wearer's vital signs (respiratory rate, heart rate) can be monitored in real time. These monitoring values ​​can be applied to the virtual scene, allowing the virtual human to have the same breathing and heartbeat sensations as a real person, making the virtual image more lifelike and thus enhancing the sense of realism in the virtual scene.

[0101] Based on the above method embodiments, this invention proposes a human feature data acquisition device, which is installed in an acquisition equipment, such as... Figure 11 As shown, it includes:

[0102] The target recognition unit 1101 is used to identify target human bodies in real time within the signal detectable area using an ultra-wideband UWB radar array, and to determine a three-dimensional model of the upper body for the identified target human body.

[0103] The detection preparation unit 1102 is used to determine the chest and abdomen region of the target human body based on the upper body three-dimensional model, and to detect the chest and abdomen region; based on the set of data points obtained from the detection, to determine the contour model of the chest and abdomen region and the monitoring control points in the contour model; the monitoring control points include respiratory monitoring control points and / or heart rate monitoring control points;

[0104] The data acquisition unit 1103 is used to perform real-time detection of the chest and abdominal region using a UWB radar array, and to determine the respiratory rate and / or heart rate of the target human body based on the detected coordinate changes of the monitoring and control points.

[0105] It should be noted that the above-described methods and apparatus embodiments are based on the same inventive concept. Since the methods and apparatus solve problems in similar ways, the implementation of the apparatus and methods can refer to each other, and repeated parts will not be described again.

[0106] Corresponding to the above method embodiments, this application also proposes a human feature data acquisition device, including a processor and a memory; the memory stores an application program that can be executed by the processor, for causing the processor to execute the human feature data acquisition method as described above. Specifically, a system or device equipped with a storage medium can be provided, on which software program code implementing the functions of any of the above embodiments is stored, and the computer (or CPU or MPU) of the system or device can read and execute the program code stored in the storage medium. In addition, the operating system or the like operating on the computer can be used to perform some or all of the actual operations through instructions based on the program code. The program code read from the storage medium can also be written to a memory set in an expansion board inserted into the computer or to a memory set in an expansion unit connected to the computer, and then the CPU or the like installed on the expansion board or expansion unit can be used to execute some or all of the actual operations based on the instructions of the program code, thereby realizing the functions of any of the above embodiments of the human feature data acquisition method.

[0107] Specifically, the memory can be implemented as various storage media such as electrically erasable programmable read-only memory (EEPROM), flash memory, and programmable programmable read-only memory (PROM). The processor can be implemented as one or more central processing units (CPUs) or one or more field-programmable gate arrays (FPGAs), wherein the FPGA integrates one or more CPU cores. Specifically, the CPU or CPU core can be implemented as a CPU or an MCU.

[0108] This application embodiment implements a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the human feature data acquisition method described above.

[0109] It should be noted that not all steps and modules in the above processes and structural diagrams are mandatory; some steps or modules can be omitted as needed. The execution order of the steps is not fixed and can be adjusted as required. The division of modules is merely for the convenience of description and functional division. In actual implementation, a module can be implemented by multiple modules, and the functions of multiple modules can also be implemented by the same module. These modules can be located in the same device or in different devices.

[0110] The hardware modules in each embodiment can be implemented mechanically or electronically. For example, a hardware module may include specially designed permanent circuitry or logic devices (such as dedicated processors, such as FPGAs or ASICs) to perform specific operations. A hardware module may also include programmable logic devices or circuitry (such as general-purpose processors or other programmable processors) temporarily configured by software to perform specific operations. The choice between mechanical implementation, dedicated permanent circuitry, or temporarily configured circuitry (such as software-configured circuitry) can be made based on cost and time considerations.

[0111] In this document, "illustrative" means "serving as an example, illustration, or description," and any illustration or embodiment described herein as "illustrative" should not be construed as a preferred or more advantageous technical solution. For the sake of brevity, each figure only schematically shows the parts relevant to the invention and does not represent their actual structure as a product. Furthermore, for the sake of clarity and ease of understanding, in some figures, components with the same structure or function are only schematically depicted, or only one is labeled. In this document, "a" does not mean that the number of relevant parts of the invention is limited to "only one," and "a" does not exclude the possibility that the number of relevant parts of the invention is "more than one." In this document, terms such as "upper," "lower," "front," "rear," "left," "right," "inner," and "outer" are used only to indicate the relative positional relationship between relevant parts, and not to limit the absolute position of these relevant parts.

[0112] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method of collecting biometric data, characterized by, The method comprises the following steps: The acquisition device uses an ultra-wideband (UWB) radar array to identify a target human body in a signal detectable area in real time, and determines an upper body three-dimensional model of the target human body; the UWB radar array is used to send an UWB linear frequency modulation continuous wave signal to the target human body to obtain the upper body three-dimensional model during the determination of the upper body three-dimensional model; Based on the upper body three-dimensional model, a chest and abdomen area of the target human body is determined, and the chest and abdomen area is detected; based on a set of data points obtained through the detection, a contour model of the chest and abdomen area and a monitoring control point in the contour model are determined; the monitoring control point comprises a breathing monitoring control point and / or a heartbeat monitoring control point; The UWB radar array is used to detect the chest and abdomen area in real time, and based on coordinate changes of the monitoring control point detected, a breathing frequency and / or a heartbeat frequency of the target human body are determined; when the target human body is in a motion state, the determination of the breathing frequency and / or the heartbeat frequency of the target human body comprises: for each coordinate data of the monitoring control point detected in a specified time range, based on three-dimensional rotation data monitored by a motion sensor of the acquisition device in real time, deviation compensation processing is performed on the coordinate data to eliminate position deviation caused by relative motion between the target human body and the acquisition device; The three-dimensional rotation data comprises three-dimensional angles and three-dimensional displacements of rotation of the acquisition device relative to the target human body; based on the coordinate data after the deviation compensation processing in the specified time range, a reciprocating movement frequency of the monitoring control point in the time range is determined, and a corresponding breathing frequency and / or heartbeat frequency are obtained.

2. The method of claim 1, wherein, The method for determining the upper body three-dimensional model of the target human body comprises the following steps: Based on a position of the target human body, a current key detection area is determined; In the key detection area, a UWB radar array is used to send a signal to the target human body, and an upper body three-dimensional model of the target human body is determined; the upper body three-dimensional model comprises upper limb part skeletal points and human upper limb template data.

3. The method of claim 2, wherein, The method for determining the upper body three-dimensional model of the target human body comprises the following steps: In the key detection area, a UWB radar array is used to send an UWB linear frequency modulation continuous wave signal to the human body; Based on echoes of the UWB linear frequency modulation continuous wave signal reflected by the target human body, a mean value method is used to remove fixed clutter in each distance unit of an echo receiving matrix; Based on the distance unit after the removal of the fixed clutter, a constant false alarm rate (CFAR) detection technology is used to determine a distance unit where the target human body is located, and based on three-dimensional array echo data of a position of the determined distance unit, three-dimensional distance and position receiving matrix data are generated; Based on the three-dimensional distance and position receiving matrix data, data preprocessing for separating different parts of the target human body is performed, and upper limb part skeletal points and human upper limb template data of the target human body are obtained.

4. The method of claim 1, wherein, The method for determining the contour model of the chest and abdomen area comprises the following steps: The data points in the data point set are subjected to thin plate spline (TPS) bending smoothing processing to obtain a contour model of the chest and abdomen region.

5. The method of claim 1, wherein, The determination of the monitoring control points comprises: Based on the respiration detection region and / or the heartbeat detection region in the preset standard human template, the respiration monitoring control points and / or the heartbeat monitoring control points are determined.

6. The method of claim 1, wherein, When the target human body is in a stationary state, the determination of the respiration frequency and / or the heartbeat frequency of the target human body comprises: According to the coordinate data of the monitoring control points detected within a specified time range, the reciprocal movement frequency of the monitoring control points within the time range is determined to obtain the corresponding respiration frequency and / or heartbeat frequency.

7. The method of claim 1, wherein, The method further comprises: During real-time detection of the chest and abdomen region, integrity checking is performed on the data point set corresponding to the detected chest and abdomen region, and when data points are missing, an alarm is triggered; When the number of alarms reaches a preset threshold within a preset time range, a matching model is searched from a preset model library based on the data point set currently detected within the time range, and the data point set within the time range is completed and filled using the searched model to obtain a corresponding complete data point set.

8. A human feature data acquisition apparatus characterized by comprising: Comprise: A target recognition unit is configured to recognize a target human body in a signal detectable region in real time by using an ultra-wideband (UWB) radar array, and determine an upper body three-dimensional model for the recognized target human body; the UWB radar array is used to send an UWB linear frequency modulation continuous wave signal to the target human body to obtain the upper body three-dimensional model during the determination of the upper body three-dimensional model; A detection preparation unit is configured to determine a chest and abdomen region of the target human body based on the upper body three-dimensional model, and detect the chest and abdomen region; Based on a data point set obtained through detection, a contour model of the chest and abdomen region and monitoring control points in the contour model are determined; the monitoring control points comprise respiration monitoring control points and / or heartbeat monitoring control points; A data acquisition unit is configured to detect the chest and abdomen region in real time by using the UWB radar array, and determine a respiration frequency and / or a heartbeat frequency of the target human body based on coordinate changes of the monitoring control points detected; when the target human body is in a motion state, the determination of the respiration frequency and / or the heartbeat frequency of the target human body comprises: for each coordinate data of the monitoring control points detected within a specified time range, three-dimensional rotation data monitored in real time by a motion sensor of an acquisition device are used to perform deviation compensation processing on the coordinate data to eliminate position deviation caused by relative motion between the target human body and the acquisition device; The three-dimensional rotation data comprise three-dimensional angles and three-dimensional displacements of rotation of the acquisition device relative to the target human body; according to the coordinate data after the deviation compensation processing within a specified time range, the reciprocal movement frequency of the monitoring control points within the time range is determined to obtain the corresponding respiration frequency and / or heartbeat frequency.

9. A human feature data acquisition apparatus characterized by comprising: Comprise a processor and a memory; The memory stores an application executable by the processor, for causing the processor to execute the human feature data acquisition method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable instructions are stored in the memory, and the computer readable instructions are used to execute the human feature data acquisition according to any one of claims 1 to 7.

11. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instructions are executed by the processor to implement the steps of the human feature data acquisition according to any one of claims 1 to 7.

Images