Human body heart rate detection device and method based on millimeter wave radar

Abstract

The invention discloses a human body heart rate detection device and method based on millimeter wave radar, and relates to the field of human body heart rate detection, and the device comprises a visual detection unit, a millimeter wave radar detection unit and a heart rate analysis module. The invention can provide a solution for remotely detecting the heart rate of the human body by adopting the millimeter-wave radar detection unit, firstly, the human body feature position is determined through the visual detection unit, and the position of the human body in a static state can be determined through a visual analysis algorithm on the basis of a pre-constructed electronic map of the current area; finally, the human body judged to be in a static state serves as a target of remote heart rate detection, the millimeter wave radar can be aligned to the heart position of the human body by controlling the action of the angle adjusting mechanism through the radar controller, and the fluctuation of the chest due to heartbeat is detected based on millimeter waves; and finally, analyzing to obtain heart rate data of the corresponding human body. The heart rates of a plurality of human body targets in the target area can be detected.

Description

Technical Field

[0001] The present invention relates to the field of human heart rate detection, and in particular to a human heart rate detection device and method based on millimeter wave radar. Background Art

[0002] At present, human heart rate detection technologies mainly include the following:

[0003] 1) Photoplethysmography (PPG) technology: It obtains heart rate information by detecting changes in blood volume on the skin surface and is widely used in wearable devices such as smart bracelets and watches.

[0004] 2) Electrocardiogram (ECG) technology: Electrodes are attached to the skin to detect the electrical activity of the heart to obtain heart rate information, mainly used in the medical field.

[0005] 3) Chest strap heart rate monitoring technology: Through the chest strap sensor, it detects chest movement and heartbeat signals, which is suitable for sports scenes.

[0006] The defects of the above different technical routes are:

[0007] 1) Photoplethysmography (PPG) technology is easily affected by factors such as skin color and ambient light, and its accuracy is limited. It needs to fit closely to the skin, and the user experience is poor.

[0008] 2) Electrocardiogram (ECG) technology requires sticking electrodes, which is inconvenient to use; it cannot achieve long-term continuous monitoring.

[0009] 3) Chest strap heart rate monitoring technology has the disadvantages of being uncomfortable to wear, being easily disturbed by movement, and having low accuracy.

[0010] Moreover, the above-mentioned technical routes can only realize one-to-one heart rate monitoring at close range, and cannot realize heart rate monitoring of multiple human bodies within an area under a specified scenario. Summary of the invention

[0011] In order to solve the above technical problems, the present invention provides a human heart rate detection device and method based on millimeter wave radar. The following technical solutions are adopted:

[0012] A human heart rate detection device based on millimeter wave radar, characterized in that it includes a visual detection unit, a millimeter wave radar detection unit and a heart rate analysis module, the visual detection unit includes multiple cameras and a visual analysis module, the multiple cameras shoot visual images within a target range, the visual analysis module is respectively connected to the multiple cameras for communication, the visual images within the target range are exchanged, and multiple human feature positions and motion data are obtained based on a human motion tracking algorithm, and when it is judged based on the motion data that the corresponding human body is in a stationary state when the amount of motion of the corresponding human body is lower than a set value within a set time, the corresponding human body is judged to be in a stationary state, and the human body position data in the stationary state is obtained;

[0013] The millimeter-wave radar detection unit includes an angle adjustment mechanism, a millimeter-wave radar and a chip-based radar controller. The base of the angle adjustment mechanism is installed on one side of the camera, and the millimeter-wave radar is installed on the movable part of the angle adjustment mechanism. The radar controller communicates with the visual analysis module to exchange the human body position data in a static state, and controls the execution actions of the angle adjustment mechanism and the millimeter-wave radar respectively, so that the millimeter-wave radar faces the human body position in a static state and emits millimeter radar waves. The heart rate analysis module communicates and interacts with the millimeter-wave radar to analyze and obtain the heart rate data of the corresponding human body.

[0014] By adopting the above technical solution, a solution for long-distance detection of human heart rate can be solved by adopting a millimeter-wave radar detection unit. Its applicable scenarios are generally relatively closed places such as wards and prisons. The heart rates of multiple human bodies in the area can be monitored without contacting the device. In order to achieve more accurate monitoring results, it is necessary to first determine the human feature position through a visual detection unit, and the human body needs to be in a relatively static state to use the millimeter-wave radar to obtain a more accurate heart rate signal. Specifically, the visual detection unit can be based on the human motion tracking algorithm to obtain multiple human feature positions and motion data. The human tracking algorithm is a relatively mature technical solution in the field of visual monitoring, which can realize automatic recognition of human features, and can also realize the calculation of displacement through a visual displacement algorithm. Finally, based on the set static state judgment standard, the corresponding human feature is in a static state. For example, the displacement can be set to be lower than a certain amount within a few seconds to realize the judgment that the human body is relatively static, and the visual analysis algorithm can be used to determine the position of the static human body based on the pre-constructed electronic map of the current area;

[0015] Finally, the human body that is judged to be in a stationary state is taken as the target of long-distance heart rate detection. The radar controller controls the action of the angle adjustment mechanism to align the millimeter-wave radar with the human heart position, and the rise and fall of the chest due to the heartbeat can be detected based on the millimeter wave, and the heart rate data of the corresponding human body can be obtained based on and finally analyzed;

[0016] Long-distance heart rate monitoring within the target area is realized, and the heart rates of multiple human targets in the target area can be detected.

[0017] Optionally, the visual analysis module includes a memory and a visual analysis chip, the memory is communicatively connected to multiple cameras respectively, the visual analysis chip is communicatively connected and interactively connected to the memory, the visual analysis chip obtains multiple human feature positions and motion data based on a human motion tracking algorithm, and determines whether there are human features in a static state, and stores the human feature position data determined to be in a static state in the memory.

[0018] By adopting the above technical solution, the visual analysis module can be implemented in the form of a memory and a visual analysis chip, or in the form of other intelligent entities with visual analysis capabilities, such as a visual analysis computer.

[0019] Optionally, the angle adjustment mechanism is an electric pan / tilt.

[0020] By adopting the above technical solution, the electric pan-tilt head can realize automatic adjustment of the radar angle and can accurately point to the target direction. In specific applications, it can also be used in conjunction with an electric linear slide or a planar motor. The electric linear slide or the planar motor can be installed on the roof of the regional target, and the angle adjustment mechanism can be installed on the active part of the electric linear slide or the planar motor. This can increase the coverage range of the millimeter-wave radar installed on the angle adjustment mechanism and avoid radar interference due to angle reasons.

[0021] Optionally, the radar controller includes a buffer, an instruction generating chip and a control chip. The buffer communicates with the memory to exchange human body feature position data in a static state. The instruction generating chip is communicatively connected to the buffer to generate control instructions for controlling the angle adjustment mechanism and the millimeter-wave radar based on the human body feature position data. The control chip is connected to the buffer to control the execution actions of the angle adjustment mechanism and the millimeter-wave radar based on the control instructions.

[0022] By adopting the above technical solution, the instruction generation chip can generate control instructions for controlling the angle adjustment mechanism and the millimeter-wave radar respectively according to the human body feature position data, and then control the angle adjustment mechanism through the control chip to realize the millimeter-wave radar to aim at the specified position of the target human body, and control the millimeter-wave radar to emit millimeter waves and receive reflected millimeter-wave signals to realize the detection of micro-displacement, thereby realizing the heart rate-related signal data that needs to be detected.

[0023] Optionally, the heart rate analysis module includes a radar data storage device and a radar data analysis computer, wherein the radar data storage device communicates with the millimeter wave radar to exchange radar data, and the radar data analysis computer detects and analyzes tiny regular movements of the human chest based on the Doppler effect, and obtains heart rate information from the micro-Doppler signal through a spectrum analysis method.

[0024] By adopting the above technical solution, the heart rate analysis module is implemented based on the radar data analysis computer. The radar data analysis computer is a computer with a certain computing power and can efficiently process the radar data.

[0025] Optionally, the heart rate analysis module also includes a display, which is communicatively connected to the radar data analysis computer to display the human body outline and corresponding heart rate data.

[0026] By adopting the above technical solution, the display can cooperate to realize richer heart rate data display. For example, it can collect human feature data of the visual detection unit, and then display the corresponding heart rate data after conventional contour marking, so that management personnel, such as ward doctors, can more conveniently understand the heart rate status of multiple people in the area.

[0027] The human heart rate detection method based on millimeter wave radar uses a human heart rate detection device based on millimeter wave radar to detect the heart rates of multiple human targets within a regional target range, including the following steps:

[0028] Step 1: multiple cameras capture visual images within the target range, and the visual image with a complete human target and the largest pixel is taken as the visual image to be analyzed;

[0029] Step 2: The visual analysis chip analyzes the visual image to be analyzed based on the human motion tracking algorithm to obtain multiple human feature positions and motion data, and determines whether the human features are in a static state based on a static state judgment condition, where the static state judgment condition is that the amount of motion within a set duration is lower than a set value;

[0030] Step 3, when it is determined that the corresponding human body feature is in a stationary state, the corresponding human body feature position data is marked based on the electronic map of the regional target;

[0031] Step 4, the buffer communicates with the memory to exchange the human body feature position data in a static state, the instruction generation chip generates control instructions for controlling the angle adjustment mechanism and the millimeter wave radar based on the human body feature position data, and the control chip controls the execution actions of the angle adjustment mechanism and the millimeter wave radar based on the control instructions, so that the millimeter wave radar points to the chest position of the human body feature in the static state, and turns on the millimeter wave radar to transmit millimeter waves;

[0032] Step 5, the radar data storage device stores the millimeter wave data, and the radar data analysis computer calculates the corresponding human heart rate data based on the spectrum analysis method;

[0033] Step 6: The display interacts with the radar data analysis computer to obtain the heart rate data of the corresponding human body, and interacts with the visual detection unit to obtain the human body feature data, and displays the contour image and heart rate data of the corresponding human body based on the real scene image.

[0034] Optionally, in step 2, the static state judgment condition is: the displacement of the geometric center of the set human body feature within five seconds is less than ten millimeters.

[0035] By adopting the above technical solution, since multiple cameras are used, it is possible that the images captured by multiple cameras all have complete human feature images. In this case, when selecting the visual image to be analyzed, the visual image with complete human target and the largest pixel should be selected as the visual image to be analyzed, which can better realize human position tracking;

[0036] The human motion tracking algorithm can track human features. The specific human motion tracking algorithm uses background subtraction, optical flow or deep learning algorithm to detect and track human motion.

[0037] For deep learning methods, convolutional neural network (CNN) is used to identify human features;

[0038] For each detected human feature, its position change in consecutive frames is calculated.

[0039] If the total amount of movement within 5 seconds is less than 10mm, it is judged that the human body feature is in a static state. At this time, the human body is generally in a relatively static state, and millimeter waves can be used to collect the heart rate signal at the chest position of the human body.

[0040] Optionally, calculating the heart rate data of the corresponding human body in step 5 includes the following sub-steps:

[0041] Step 51, amplifying, filtering and down-converting the radar data;

[0042] Step 52, detecting micro-movement of the chest by analyzing phase change or frequency change of the echo signal;

[0043] Step 53, separating the motion signal associated with the heart beat from the received micro-motion signal as the heart rate signal;

[0044] Step 54, extract the frequency component in the heart rate signal through Fourier transform, the formula is as follows:

[0045] Xf=Fx(t);

[0046] Where Xf is the Fourier transform of the signal x(t), and F· represents the Fourier transform operation;

[0047] Step 55, find the peak value of the main frequency in the Fourier transform result and convert it into heart rate.

[0048] Optionally, in step 55, the formula for converting the peak value of the main frequency into the heart rate is:

[0049]

[0050] Where HR is the heart rate, f peak is the peak value of the main frequency in the heart rate signal, f sampleis the sampling frequency.

[0051] By adopting the above technical solutions, millimeter wave radars are usually selected in the industrial, scientific and medical (ISM) bands of 24GHz or 77GHz, which have good penetration into the human body and will not cause harm. It is also possible to design a dedicated millimeter wave radar receiving antenna array, design a suitable antenna array to focus on the chest area, and improve the signal-to-noise ratio. Of course, it is also feasible to directly use millimeter wave radars for transmission and reception. Millimeter wave radars can detect human heart rate without contact, and are suitable for medical monitoring, health monitoring and other fields.

[0052] In summary, the present invention includes at least one of the following beneficial technical effects:

[0053] The present invention can provide a human heart rate detection device and method based on millimeter wave radar. By adopting a millimeter wave radar detection unit to solve the solution of long-distance detection of human heart rate, the heart rates of multiple human bodies in a regional place can be monitored without contacting the device. In order to achieve a more accurate monitoring result, it is necessary to first determine the characteristic position of the human body through a visual detection unit, and the human body needs to be in a relatively static state to adopt the millimeter wave radar to obtain a more accurate heart rate signal. The visual analysis algorithm can be used to determine the position of the static human body based on a pre-built electronic map of the current area.

[0054] Finally, the human body that is judged to be in a stationary state is taken as the target of long-distance heart rate detection. The radar controller controls the action of the angle adjustment mechanism to align the millimeter-wave radar with the human heart position, and the rise and fall of the chest due to the heartbeat can be detected based on the millimeter wave, and the heart rate data of the corresponding human body can be obtained based on and finally analyzed;

[0055] Long-distance heart rate monitoring within the target area is realized, and the heart rates of multiple human targets in the target area can be detected. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of the connection principle of electrical components of the human heart rate detection device based on millimeter wave radar of the present invention;

[0057] Figure 2 It is a schematic diagram of the principle of detecting the human heart rate state by the human heart rate detection device based on millimeter wave radar of the present invention;

[0058] Figure 3 The present invention is a schematic diagram of a display of a human heart rate detection device based on millimeter wave radar showing a human heart rate.

[0059] Explanation of the accompanying drawings: 11. Camera; 12. Visual analysis module; 121. Memory; 122. Visual analysis chip; 21. Angle adjustment mechanism; 22. Millimeter wave radar; 23. Radar controller; 231. Cache; 232. Instruction generation chip; 233. Control chip; 31. Radar data storage; 32. Radar data analysis computer; 33. Display. DETAILED DESCRIPTION

[0060] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0061] The embodiments of the present invention disclose a human heart rate detection device and method based on millimeter wave radar.

[0062] Reference Figure 1-Figure 3 Embodiment 1, a human heart rate detection device based on millimeter wave radar, characterized in that: it includes a visual detection unit, a millimeter wave radar detection unit and a heart rate analysis module, the visual detection unit includes multiple cameras 11 and a visual analysis module 12, the multiple cameras 11 shoot visual images within the target range, the visual analysis module 12 is respectively connected to the multiple cameras 11 for communication, the visual images within the target range are exchanged, and multiple human feature positions and motion data are obtained based on a human motion tracking algorithm, when it is judged based on the motion data that the corresponding human body is in a stationary state when the amount of motion within the set time is lower than a set value, the corresponding human body is judged to be in a stationary state, and the human body position data in the stationary state is obtained;

[0063] The millimeter-wave radar detection unit includes an angle adjustment mechanism 21, a millimeter-wave radar 22 and a chip-based radar controller 23. The base of the angle adjustment mechanism 21 is installed on one side of the camera 11, and the millimeter-wave radar 22 is installed on the movable part of the angle adjustment mechanism 21. The radar controller 23 communicates with the visual analysis module 12 to exchange the human body position data in a static state, and controls the execution actions of the angle adjustment mechanism 21 and the millimeter-wave radar 22 respectively, so that the millimeter-wave radar 22 faces the human body position in a static state and emits millimeter radar waves. The heart rate analysis module communicates and interacts with the millimeter-wave radar 22 to analyze and obtain the heart rate data of the corresponding human body.

[0064] The solution of long-distance detection of human heart rate can be solved by adopting millimeter-wave radar detection unit. Its applicable scenarios are generally relatively closed places such as wards and prisons. The heart rates of multiple human bodies in the area can be monitored without contact equipment. In order to achieve more accurate monitoring results, it is necessary to first determine the position of human features through the visual detection unit, and the human body needs to be in a relatively static state to use millimeter-wave radar to obtain more accurate heart rate signals. Specifically, the visual detection unit can obtain multiple human feature positions and motion data based on the human motion tracking algorithm. The human tracking algorithm is a relatively mature technical solution in the field of visual monitoring. It can realize automatic recognition of human features, and can also calculate the displacement through the visual displacement algorithm. Finally, based on the set static state judgment standard, the corresponding human feature is in a static state. For example, the displacement can be set to be lower than a certain amount within a few seconds to realize the judgment that the human body is relatively static, and the visual analysis algorithm can be used to determine the position of the static human body based on the pre-built electronic map of the current area;

[0065] Finally, the human body judged to be in a stationary state is taken as the target of long-distance heart rate detection. The radar controller 23 controls the action of the angle adjustment mechanism 21 to realize the millimeter wave radar 22 to be aimed at the human heart position, and the rise and fall of the chest due to the heartbeat can be detected based on the millimeter wave, and the heart rate data of the corresponding human body can be obtained based on and finally analyzed;

[0066] Long-distance heart rate monitoring within the target area is realized, and the heart rates of multiple human targets in the target area can be detected.

[0067] In embodiment 2, the visual analysis module 12 includes a memory 121 and a visual analysis chip 122. The memory 121 is respectively connected to communicate with multiple cameras 11, and the visual analysis chip 122 is connected to communicate and interact with the memory 121. The visual analysis chip 122 obtains multiple human feature positions and motion data based on a human motion tracking algorithm, and determines whether there are human features in a static state, and stores the human feature position data determined to be in a static state in the memory 121.

[0068] The visual analysis module 12 may be implemented in the form of a memory 121 and a visual analysis chip 122, or may be other intelligent entities with visual analysis capabilities, such as a visual analysis computer.

[0069] In the third embodiment, the angle adjustment mechanism 21 is an electric pan / tilt head.

[0070] The electric pan-tilt head can realize automatic adjustment of the radar angle and can accurately point to the target direction. In specific applications, it can also be used in conjunction with an electric linear slide or a planar motor. The electric linear slide or planar motor can be installed on the roof of the regional target, and the angle adjustment mechanism 21 can be installed on the active part of the electric linear slide or planar motor. This can increase the coverage range of the millimeter-wave radar 22 installed on the angle adjustment mechanism 21 and avoid radar interference due to angle reasons.

[0071] In Example 4, the radar controller 23 includes a buffer 231, an instruction generation chip 232 and a control chip 233. The buffer 231 communicates with the memory 121 to exchange the human body feature position data in a static state. The instruction generation chip 232 is communicatively connected to the buffer 231, and based on the human body feature position data, control instructions for controlling the angle adjustment mechanism 21 and the millimeter-wave radar 22 are respectively generated. The control chip 233 is connected to the buffer 231, and based on the control instructions, the execution actions of the angle adjustment mechanism 21 and the millimeter-wave radar 22 are respectively controlled.

[0072] The instruction generation chip 232 can generate control instructions for controlling the angle adjustment mechanism 21 and the millimeter-wave radar 22 respectively according to the human body feature position data, and then control the angle adjustment mechanism 21 through the control chip 233 to enable the millimeter-wave radar 22 to aim at the specified position of the target human body, and control the millimeter-wave radar 22 to emit millimeter waves and receive reflected millimeter-wave signals to detect micro-displacements, thereby realizing the heart rate-related signal data that needs to be detected.

[0073] Convert the pixel coordinates of the human feature position data into coordinates in the actual world coordinate system. You need to know the camera's intrinsic parameters (focal length, principal point, etc.) and extrinsic parameters (the position and orientation of the camera in the world coordinate system).

[0074] Calculate the target angle: According to the actual coordinates of the human chest position and the current position of the millimeter wave radar 22, calculate the angle that the millimeter wave radar 22 needs to rotate so as to aim at the human chest.

[0075] An encoder or other sensor is used to determine the current angle position of the millimeter wave radar 22. The instruction generation chip 232 generates a control signal that can be recognized by the angle adjustment mechanism 21 according to the angle conversion data.

[0076] Embodiment 5, the heart rate analysis module includes a radar data storage device 31 and a radar data analysis computer 32. The radar data storage device 31 communicates with the millimeter wave radar 22 to exchange radar data. The radar data analysis computer 32 detects and analyzes the tiny regular movements of the human chest based on the Doppler effect, and obtains the heart rate information from the micro-Doppler signal through a spectrum analysis method.

[0077] The heart rate analysis module is implemented based on the radar data analysis computer 32. The radar data analysis computer 32 is a computer with a certain computing power and can efficiently process the radar data.

[0078] In Example 6, the heart rate analysis module further includes a display 33, which is in communication with the radar data analysis computer 32 and displays the human body contour and corresponding heart rate data.

[0079] The display 33 can be used to realize richer heart rate data display. For example, it can collect human feature data from visual inspection, and then display the corresponding heart rate data after conventional contour marking, so that management personnel, such as ward doctors, can more conveniently understand the heart rate status of multiple people in the area.

[0080] Embodiment 7, a human heart rate detection method based on millimeter wave radar, uses a human heart rate detection device based on millimeter wave radar to detect the heart rates of multiple human targets within a regional target range, comprising the following steps:

[0081] Step 1, multiple cameras 11 capture visual images within the target range, and take the visual image with a complete human target and the largest pixel as the visual image to be analyzed;

[0082] Step 2, the visual analysis chip 122 analyzes the visual image to be analyzed based on the human motion tracking algorithm to obtain multiple human feature positions and motion data, and determines whether the human features are in a static state based on a static state judgment condition, where the static state judgment condition is that the amount of motion within a set duration is lower than a set value;

[0083] Step 3, when it is determined that the corresponding human body feature is in a stationary state, the corresponding human body feature position data is marked based on the electronic map of the regional target;

[0084] Step 4, the buffer 231 communicates and exchanges the human body feature position data in a static state with the memory 121, the instruction generation chip 232 generates control instructions for controlling the angle adjustment mechanism 21 and the millimeter wave radar 22 based on the human body feature position data, and the control chip 233 controls the execution actions of the angle adjustment mechanism 21 and the millimeter wave radar 22 based on the control instructions, so that the millimeter wave radar 22 points to the chest position of the human body feature in a static state, and turns on the millimeter wave radar 22 to emit millimeter waves;

[0085] Step 5, the radar data storage 31 stores the millimeter wave data, and the radar data analysis computer 3 calculates the heart rate data of the corresponding human body based on the spectrum analysis method;

[0086] Step 6, the display 33 interacts with the radar data analysis computer 3 to obtain the heart rate data of the corresponding human body, and interacts with the visual detection unit to obtain the human body feature data, and displays the contour image and heart rate data of the corresponding human body based on the real scene image.

[0087] In Example 8, in step 2, the static state judgment condition is: the displacement of the geometric center of the set human body feature within five seconds is less than ten millimeters.

[0088] Since multiple cameras 11 are used, it is possible that the images captured by multiple cameras 11 all have complete human feature images. In this case, when selecting the visual image to be analyzed, the visual image with complete human target and the largest pixel should be selected as the visual image to be analyzed, so as to better realize the human position tracking;

[0089] The human motion tracking algorithm can track human features. The specific human motion tracking algorithm uses background subtraction, optical flow or deep learning algorithm to detect and track human motion.

[0090] For deep learning methods, convolutional neural network (CNN) is used to identify human features;

[0091] For each detected human feature, its position change in consecutive frames is calculated.

[0092] If the total amount of movement within 5 seconds is less than 10mm, it is judged that the human body feature is in a static state. At this time, the human body is generally in a relatively static state, and millimeter waves can be used to collect the heart rate signal at the chest position of the human body.

[0093] In embodiment 9, the step 5 of calculating the heart rate data of the corresponding human body includes the following sub-steps:

[0094] Step 51, amplifying, filtering and down-converting the radar data;

[0095] Step 52, detecting micro-movement of the chest by analyzing phase change or frequency change of the echo signal;

[0096] Step 53, separating the motion signal associated with the heart beat from the received micro-motion signal as the heart rate signal;

[0097] Step 54, extract the frequency component in the heart rate signal through Fourier transform, the formula is as follows:

[0098] Xf=Fx(t);

[0099] Where Xf is the Fourier transform of the signal x(t), and F· represents the Fourier transform operation;

[0100] Step 55, find the peak value of the main frequency in the Fourier transform result and convert it into heart rate.

[0101] In Example 10, in step 55, the formula for converting the peak value of the main frequency into the heart rate is:

[0102]

[0103] Where HR is the heart rate, f peak is the peak value of the main frequency in the heart rate signal, f sample is the sampling frequency.

[0104] Millimeter-wave radars are usually selected in the industrial, scientific and medical (ISM) bands of 24 GHz or 77 GHz. These frequencies have good penetration into the human body and will not cause harm. It is also possible to design a dedicated millimeter-wave radar receiving antenna array and design a suitable antenna array to focus on the chest area to improve the signal-to-noise ratio. Of course, it is also feasible to directly use millimeter-wave radars for transmission and reception. Millimeter-wave radars can detect human heart rate without contact and are suitable for medical monitoring, health monitoring and other fields.

[0105] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A human heart rate detection device based on millimeter wave radar, characterized in that: The invention comprises a visual detection unit, a millimeter wave radar detection unit and a heart rate analysis module, wherein the visual detection unit comprises a plurality of cameras (11) and a visual analysis module (12), wherein the plurality of cameras (11) capture visual images within a target range, and the visual analysis module (12) is respectively connected to the plurality of cameras (11) for communicating with each other to exchange visual images within the target range, and obtains a plurality of human feature positions and motion data based on a human motion tracking algorithm, and when it is determined based on the motion data that the amount of motion of the corresponding human body within a set time is lower than a set value, it is determined that the corresponding human body is in a stationary state, and the stationary human body position data is obtained; The millimeter wave radar detection unit comprises an angle adjustment mechanism (21), a millimeter wave radar (22) and a chip-based radar controller (23); the base of the angle adjustment mechanism (21) is installed on one side of the camera (11); the millimeter wave radar (22) is installed on the movable part of the angle adjustment mechanism (21); the radar controller (23) communicates with the visual analysis module (12) to exchange the human body position data in a static state, and controls the execution actions of the angle adjustment mechanism (21) and the millimeter wave radar (22) respectively, so that the millimeter wave radar (22) faces the human body position in a static state and emits millimeter radar waves; the heart rate analysis module communicates with the millimeter wave radar (22) to analyze and obtain the heart rate data of the corresponding human body.

2. The human heart rate detection device based on millimeter wave radar according to claim 1, characterized in that: The visual analysis module (12) comprises a memory (121) and a visual analysis chip (122); the memory (121) is respectively connected to the plurality of cameras (11) in communication; the visual analysis chip (122) is connected to the memory (121) in communication and interaction; the visual analysis chip (122) obtains a plurality of human feature positions and motion data based on a human motion tracking algorithm, determines whether a human feature is in a stationary state, and stores the human feature position data determined to be in a stationary state in the memory (121).

3. The human heart rate detection device based on millimeter wave radar according to claim 2 is characterized in that: The angle adjustment mechanism (21) is an electric pan / tilt platform.

4. The human heart rate detection device based on millimeter wave radar according to claim 3 is characterized in that: The radar controller (23) comprises a buffer (231), an instruction generation chip (232) and a control chip (233); the buffer (231) communicates with a memory (121) to exchange human body feature position data in a static state; the instruction generation chip (232) is communicatively connected to the buffer (231) to respectively generate control instructions for controlling an angle adjustment mechanism (21) and a millimeter wave radar (22) based on the human body feature position data; the control chip (233) is connected to the buffer (231) to respectively control the execution actions of the angle adjustment mechanism (21) and the millimeter wave radar (22) based on the control instructions.

5. The human heart rate detection device based on millimeter wave radar according to claim 4 is characterized in that: The heart rate analysis module comprises a radar data storage (31) and a radar data analysis computer (32), wherein the radar data storage (31) communicates with the millimeter wave radar (22) to exchange radar data, and the radar data analysis computer (32) detects and analyzes the minute regular movements of the human chest based on the Doppler effect, and obtains heart rate information from the micro-Doppler signal by using a spectrum analysis method.

6. The human heart rate detection device based on millimeter wave radar according to claim 5, characterized in that: The heart rate analysis module also includes a display (33), which is in communication connection with the radar data analysis computer (32) and displays the human body contour and corresponding heart rate data.

7. A human heart rate detection method based on millimeter wave radar, characterized in that: The human heart rate detection device based on millimeter wave radar as claimed in claim 6 is used to detect the heart rates of multiple human targets within the regional target range, comprising the following steps: Step 1, multiple cameras (11) capture visual images within the target range, and select the visual image with a complete human target and the largest pixel as the visual image to be analyzed; Step 2, the visual analysis chip (122) analyzes the visual image to be analyzed based on the human motion tracking algorithm to obtain multiple human feature positions and motion data, and determines whether the human features are in a static state based on a static state judgment condition, wherein the static state judgment condition is that the amount of motion within a set duration is lower than a set value; Step 3, when it is determined that the corresponding human body feature is in a stationary state, the corresponding human body feature position data is marked based on the electronic map of the regional target; Step 4, the buffer (231) communicates with the memory (121) to exchange the human body feature position data in a static state, the instruction generation chip (232) generates control instructions for controlling the angle adjustment mechanism (21) and the millimeter wave radar (22) based on the human body feature position data, and the control chip (233) controls the execution actions of the angle adjustment mechanism (21) and the millimeter wave radar (22) based on the control instructions, so that the millimeter wave radar (22) points to the chest position of the human body feature in a static state, and the millimeter wave radar (22) is turned on to emit millimeter waves; Step 5, the radar data storage device (31) stores the millimeter wave data, and the radar data analysis computer (3) calculates the heart rate data of the corresponding human body based on the spectrum analysis method; Step 6, the display (33) interacts with the radar data analysis computer (3) to obtain the heart rate data of the corresponding human body, and interacts with the visual detection unit to obtain the human body feature data, and displays the contour image and heart rate data of the corresponding human body based on the real scene image.

8. The method for detecting human heart rate based on millimeter wave radar according to claim 7, characterized in that: In step 2, the static state judgment condition is: the displacement of the geometric center of the set human body feature within five seconds is less than ten millimeters.

9. The method for detecting human heart rate based on millimeter wave radar according to claim 8, characterized in that: The step 5 of calculating the heart rate data of the corresponding human body includes the following sub-steps: Step 51, amplifying, filtering and down-converting the radar data; Step 52, detecting micro-movement of the chest by analyzing phase change or frequency change of the echo signal; Step 53, separating the motion signal associated with the heart beat from the received micro-motion signal as the heart rate signal; Step 54, extract the frequency component in the heart rate signal through Fourier transform, the formula is as follows: X9f0=F{x(t)}; where X(f) is the Fourier transform of the signal x(t), and F{·} represents the Fourier transform operation; Step 55, find the peak value of the main frequency in the Fourier transform result and convert it into heart rate.

10. The method for detecting human heart rate based on millimeter wave radar according to claim 9, characterized in that: In step 55, the formula for converting the peak value of the main frequency into the heart rate is: Where HR is heart rate, f peak is the peak value of the main frequency in the heart rate signal, f sample is the sampling frequency.

Images