A Human Posture Recognition Method and System Based on Millimeter-Wave Radar Detection

Abstract

The present invention relates to the technical field of millimeter-wave radar detection, and specifically discloses a human body posture recognition method and system based on millimeter-wave radar detection. The method includes the following steps: S1: constructing a human body posture information database, S2: identifying target human body information, S3: analyzing the dynamic characteristics of the target human body, S4: judging the comprehensive coefficient of the dynamic characteristics of the target human body, S5: outputting the recognition result of the human body posture in the static state, and S6: outputting the recognition result of the human body posture in the motion state; by integrating advanced millimeter-wave radar technology and the human body posture information database, it is possible to achieve efficient and accurate recognition of the human body posture, providing users with high-quality posture recognition results. In addition, the present invention introduces a dynamic characteristic comprehensive coefficient DCC and a motion situation evaluation coefficient SEC, and combines preset thresholds to achieve intelligent adaptive judgment of the human body state.

Description

Technical Field

[0001] The present invention relates to the technical field of millimeter-wave radar detection, and particularly to a human body posture recognition method and system based on millimeter-wave radar detection. Background Art

[0002] A millimeter-wave radar is a sensor that uses electromagnetic waves in the frequency range of 30 GHz to 300 GHz for detection. The wavelength of this frequency band ranges from 1 millimeter to 10 millimeters, so it is called millimeter wave. Compared with traditional optical and infrared sensors, millimeter-wave radars have significant advantages, such as being not restricted by lighting conditions, being able to work in bad weather conditions, and having the ability to penetrate non-metallic materials.

[0003] In a Chinese invention application with the application publication number CN115345908B, a human body posture recognition method based on a millimeter-wave radar is disclosed. The method includes the following steps: Step 1, emitting an electromagnetic wave signal to the space range to be measured through a millimeter-wave radar, and obtaining target point traces based on the echo signal; Step 2, for the target point traces, condensing the point traces into different regions, and matching them to known tracks according to conditions to obtain the movement track of the target; Step 3, according to the coordinate positions (x, y, z) of the target point cloud in the 3D space, and the signal-to-noise ratio intensity (l) of each point; Step 4, calculating the velocity, distance, and height changes of the dynamic Doppler spectrum respectively, and extracting the movement characteristics of the dynamic Doppler spectrum; Step 5, proposing a sliding window target area detection algorithm to perform target detection within the target area; Step 6, inputting the obtained movement characteristics, the RGB of the point trace intensity, and its posture label into a neural network for training and learning for recognition.

[0004] In the above invention application, it is necessary to first obtain the movement track through complex operations such as point trace condensation and track matching according to conditions, and then calculate the relevant parameters of the dynamic Doppler spectrum after multiple steps of processing such as combining the 3D space coordinate position and signal-to-noise ratio intensity. The process is cumbersome, not only time-consuming in calculation, but also the cumulative error in the intermediate links may increase, reducing the accuracy and real-time performance of the final posture recognition; in addition, the public patent is not comprehensive enough in feature mining, lacks the utilization of key features such as the offset of the human body center of gravity, and does not mention the construction and application of a similar complete posture atlas database, increasing the risk of misjudgment in posture recognition, especially the accuracy of static posture recognition in complex scenarios is easily affected. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a human body posture recognition method and system based on millimeter-wave radar detection to solve the problems raised in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solutions: A human body posture recognition method based on millimeter-wave radar detection, comprising the following steps: S1: constructing a human body posture information database, S2: identifying target human body information, S3: analyzing the dynamic characteristics of the target human body, S4: judging the comprehensive coefficient of the dynamic characteristics of the target human body, S5: outputting the recognition result of the human body posture in the static state, and S6: outputting the recognition result of the human body posture in the motion state;

[0007] S1: Constructing a human body posture information database: Obtaining various action posture maps of the human body posture in the static or motion state to form a human body posture information database;

[0008] S2: Identifying target human body information: Transmitting an electromagnetic wave signal and receiving an echo signal within the target detection range by a millimeter-wave radar, preprocessing the echo signal, and identifying and screening out the target human body information based on the preprocessed echo signal;

[0009] S3: Analyzing the dynamic characteristics of the target human body: Detecting the displacement change amount, azimuth angle deviation degree, and speed change rate of the target human body, and analyzing to obtain the comprehensive coefficient DCC of the dynamic characteristics of the target human body based on the millimeter-wave radar echo signal parameters;

[0010] S4: Judging the comprehensive coefficient of the dynamic characteristics of the target human body: Judging based on the comprehensive coefficient of the dynamic characteristics of the target human body and a preset threshold. If it is judged that the target human body is in a static state, further execute S5; otherwise, further execute S6;

[0011] S5: Outputting the recognition result of the human body posture in the static state: Scanning the target human body in the static state, constructing a posture map of the key nodes of the human body, comparing it with the posture maps in the human body posture information database, screening out the posture maps that conform to the human body posture in the static state, and outputting the recognition result of the human body posture in the static state;

[0012] S6: Outputting the recognition result of the human body posture in the motion state: Detecting the acceleration value, center of gravity deviation rate, and intensity change rate of the echo signal when the target human body is moving, analyzing to obtain the motion trend evaluation coefficient SEC of the target human body, and comparing it with a preset motion posture threshold range to analyze the human body posture, and outputting the recognition result of the human body posture in the motion state.

[0013] Preferably, the human body posture information database includes, but is not limited to, static human body posture maps of standing, sitting upright, lying on the side, and squatting, includes, but is not limited to, motion human body posture maps of walking, running, and running up stairs, includes, but is not limited to, human body posture maps of key parts and key nodes of the human body, and includes, but is not limited to, metadata related to action labels, timestamps, and postures.

[0014] Preferably, the target detection range is specifically the living space of the target human body. The living space of the target human body is determined as the target detection range, and the millimeter-wave radar is used to identify the human body posture within the target detection range.

[0015] Preferably, the specific content of the dynamic feature analysis of the target human body is as follows:

[0016] S31: Obtain the first distance D1 between the target human body and the millimeter-wave radar and the second distance D2 after a time Δt, and substitute them into the formula to obtain the displacement change ΔD of the target human body;

[0017] Obtain the azimuth angle θ of the target human body and substitute it into the formula to obtain the azimuth angle offset Ao of the target human body, where θ0 represents the initial azimuth angle of the target human body;

[0018] Obtain the first speed V1 of the target human body and the second speed V2 after a time Δt, and substitute them into the formula to obtain the speed change rate ΔV of the target human body;

[0019] S32: Obtain the millimeter-wave radar echo signal parameters, and the echo signal parameters include the intensity St, the fundamental frequency f, and the phase of the millimeter-wave radar echo signal

[0020] S33: Extract the displacement change ΔD of the target human body, the azimuth angle offset Ao of the target human body, the speed change rate ΔV of the target human body, the intensity St of the echo signal, the fundamental frequency f, and the phase Substitute them into the formula respectively to obtain the dynamic feature comprehensive coefficient DCC of the target human body, where e is the natural constant and α is the error factor.

[0021] Preferably, the specific content of the judgment on the dynamic feature comprehensive coefficient of the target human body is as follows:

[0022] Extract the dynamic feature comprehensive coefficient of the target human body and compare it with the preset threshold. If the dynamic feature comprehensive coefficient of the target human body is less than the preset threshold, it is judged that the target human body is in a static state, and S5 is further executed. If the dynamic feature comprehensive coefficient of the target human body is greater than the preset threshold, it is judged that the target human body is in a moving state, and S6 is further executed.

[0023] Preferably, the specific content of the output of the static state human body posture recognition result is as follows:

[0024] Based on the judgment result of S4, use an infrared sensor to comprehensively scan the target human body in a static state, and extract the key nodes of the human body from the scanned data;

[0025] A three-dimensional space coordinate system is established, which is represented by three axes XYZ in a rectangular coordinate system. Among them, the X-axis is in the vertical direction, that is, the standing direction of the target human body, the Y-axis is in the front-back direction of the target human body, and the Z-axis is in the left-right direction of the target human body. The key nodes of the human body extracted from the scan data are projected into the three-dimensional space coordinate system to obtain the coordinate values of the key nodes. Based on the coordinate values, the connection relationship and relative positions between the key nodes are depicted, and the posture map of the target human body is constructed;

[0026] The constructed posture map of the target human body is compared with the maps in the pre-constructed human body posture information database. In the comparison results, the target maps that match the static-state human body posture maps in the database are screened out, and the similarity between the posture map of the target human body and the static-state maps in the database is compared with a preset similarity threshold. When the similarity between the posture map of the target human body and the static-state maps in the database is greater than the preset similarity threshold, it is determined that the target map conforms to the static state, and the recognition result of the static-state human body posture is output.

[0027] Preferably, the calculation steps of the motion trend evaluation coefficient of the target human body are specifically as follows:

[0028] S61: Calculate the acceleration value when the target human body is moving. Taking the ground as the reference, a space coordinate system 0XYZ is established, and the coordinate system of the target human body during movement is OA X A Y A Z , and through rotating β degrees around the X-axis, rotating γ degrees around the Y-axis, and rotating α around the Z-axis for coordinate system conversion, the coordinate calculation formula is as follows:

[0029]

[0030] , where, A X 、A Y 、A Z respectively represent the acceleration values of the target human body on the X, Y, and Z axes during movement;

[0031] S62: Obtain the initial coordinates of the human body's center of gravity as (x0, y0, z0). After time t′, the center of gravity coordinates become (x1, y1, z1), and substitute them into the formula to obtain the center of gravity offset rate Gsr of the target human body, where L is the height of the target human body;

[0032] S63: Obtain the echo signal intensity S(t) received at time t. After the time interval [t0, t1], the echo signal intensity changes from S(t0) to S(t1), and substitute them into the formula to obtain the echo signal intensity change rate Sr of the target human body;

[0033] S63: Calculate the motion state evaluation coefficient SEC of the target human body. The specific calculation model is as follows:

[0034] where e represents the natural constant, and μ represents the angle between the radar beam and the human body's motion direction.

[0035] Preferably, the specific content output by the motion state human body posture recognition result is as follows:

[0036] Extract the motion state evaluation coefficient of the target human body, and use the clustering analysis method to compare the motion state evaluation coefficient of the target human body with the preset motion posture threshold interval one by one, and output the human body posture recognition results corresponding to the preset motion posture threshold intervals respectively.

[0037] To achieve the above object, the present invention provides the following technical solution: A human body posture recognition system based on millimeter-wave radar detection, implementing the above-mentioned human body posture recognition method based on millimeter-wave radar detection, includes:

[0038] Construct a human body posture information database module: Obtain various action posture maps of the human body posture in a static or moving state, and form a human body posture information database;

[0039] Identify the target human body information module: Transmit an electromagnetic wave signal within the target detection range through a millimeter-wave radar and receive the echo signal, and preprocess the echo signal, and identify and screen out the target human body information based on the preprocessed echo signal;

[0040] Dynamic feature analysis module of the target human body: Detect the displacement change amount, azimuth deviation degree, and speed change rate of the target human body, and analyze and obtain the dynamic feature comprehensive coefficient DCC of the target human body based on the millimeter-wave radar echo signal parameters;

[0041] Dynamic feature comprehensive coefficient judgment module of the target human body: Judge based on the dynamic feature comprehensive coefficient of the target human body and the preset threshold. If it is judged that the target human body is in a static state, further execute the static state human body posture recognition result output module, otherwise further execute the dynamic state human body posture recognition result output module;

[0042] Static state human body posture recognition result output module: Scan the target human body in a static state, construct a posture map of the key nodes of the human body, compare it with the posture maps in the human body posture information database, and screen out the posture maps that conform to the static state human body posture, and output the static state human body posture recognition results;

[0043] Moving state human body posture recognition result output module: Detect the acceleration value, center of gravity offset rate, and intensity change rate of the echo signal when the target human body is moving, analyze to obtain the motion trend evaluation coefficient SEC of the target human body, and compare it with the preset motion posture threshold interval to analyze the human body posture, and output the recognition result of the moving state human body posture.

[0044] Technical effects and advantages of the present invention:

[0045] 1. By detecting and analyzing multi-dimensional information such as the displacement change amount, azimuth angle deviation degree, and speed change rate of the target human body, the present invention can accurately capture the tiny dynamic changes of the human body, and calculate the dynamic feature comprehensive coefficient DCC based on this, providing more refined data support for posture recognition;

[0046] 2. By comparing the dynamic feature comprehensive coefficient DCC of the target human body with the preset threshold, the present invention can automatically determine whether the target human body is in a static state or a moving state, and select the corresponding posture recognition strategy accordingly, realizing intelligent and adaptive recognition of human body postures. Description of the drawings

[0047] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to the following drawings without creative efforts.

[0048] Figure 1 It is a schematic flowchart of a human body posture recognition method based on millimeter-wave radar detection of the present invention.

[0049] Figure 2 It is a schematic diagram of the modules of a human body posture recognition system based on millimeter-wave radar detection of the present invention. Specific embodiments

[0050] 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 only a part of the embodiments of the present invention, rather than all 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.

[0051] Embodiment 1

[0052] Please refer to Figure 1As shown in the figure, the present invention provides a human body posture recognition method based on millimeter-wave radar detection, including the following steps: S1: constructing a human body posture information database, S2: identifying target human body information, S3: analyzing the dynamic characteristics of the target human body, S4: judging the comprehensive coefficient of the dynamic characteristics of the target human body, S5: outputting the recognition result of the static state human body posture, and S6: outputting the recognition result of the moving state human body posture;

[0053] S1: Constructing a human body posture information database: Obtain various action posture maps of the human body in a static or moving state, and form a human body posture information database;

[0054] In this embodiment, it should be specifically noted that the human body posture information database includes, but is not limited to, static human body posture maps of standing, sitting upright, lying on the side, and squatting, including, but is not limited to, moving human body posture maps of walking, running, and running up stairs, including, but is not limited to, human body posture maps of key parts and key nodes of the human body, including, but is not limited to, action labels, timestamps, and metadata related to postures.

[0055] S2: Identifying target human body information: Transmit electromagnetic wave signals in the target detection range through a millimeter-wave radar and receive echo signals, and preprocess the echo signals, and identify and screen out target human body information based on the preprocessed echo signals;

[0056] In this embodiment, it should be specifically noted that the target detection range is specifically the living space of the target human body. The living space of the target human body is determined as the target detection range, and the human body posture is recognized within the target detection range through a millimeter-wave radar.

[0057] In this embodiment, it should be specifically noted that the specific content of identifying the target human body information is:

[0058] Using the convolutional neural network (CNN) in deep learning technology, convert the echo signals received by the millimeter-wave radar into two-dimensional image data, where the distance and angle information are mapped to the coordinates of the image, and the echo intensity is used as the pixel value; By constructing a CNN model including convolutional layers, pooling layers, and fully connected layers, and training with a large amount of millimeter-wave radar data containing human and non-human targets, the model can automatically learn and extract complex features related to the human body in the radar signals, and finally achieve accurate recognition and classification of the target human body.

[0059] S3: Analyzing the dynamic characteristics of the target human body: Detect the displacement change amount, azimuth angle deviation degree, and speed change rate of the target human body, and analyze and obtain the dynamic characteristic comprehensive coefficient DCC based on the millimeter-wave radar echo signal parameters;

[0060] In this embodiment, it should be specifically noted that the specific content of the dynamic characteristic analysis of the target human body is:

[0061] S31: Obtain the first distance D1 between the target human body and the millimeter-wave radar, and the second distance D2 after a time Δt, and substitute them into the formula to obtain the displacement change amount ΔD of the target human body;

[0062] Obtain the azimuth angle θ of the target human body, and substitute it into the formula to obtain the azimuth angle offset Ao of the target human body, where θ0 represents the initial azimuth angle of the target human body;

[0063] Obtain the first speed V1 of the target human body and the second speed V2 after a time Δt, and substitute them into the formula to obtain the speed change rate ΔV of the target human body;

[0064] S32: Obtain the millimeter-wave radar echo signal parameters, where the echo signal parameters include the intensity St, the fundamental frequency f, and the phase of the millimeter-wave radar echo signal

[0065] In this embodiment, it should be specifically noted that the intensity, frequency, and phase of the radar echo signal can be obtained directly through the radar system; the intensity information is obtained by measuring the amplitude of the received signal; the fundamental frequency can be directly obtained from the radar system; and the phase information needs to be obtained through a phase detector and a coherent processing method.

[0066] S33: Extract the displacement change amount ΔD of the target human body, the azimuth angle offset Ao of the target human body, the speed change rate ΔV of the target human body, the intensity St of the echo signal, the fundamental frequency f, and the phase Substitute them into the formula to obtain the dynamic characteristic comprehensive coefficient DCC of the target human body, where e is the natural constant and α is the error factor.

[0067] In this embodiment, it should be specifically noted that the specific obtaining methods for the distance between the target human body and the millimeter-wave radar, the azimuth angle of the target human body, and the speed of the target human body are as follows:

[0068] Obtain the received signal frequency f2 and the transmitted signal frequency f1 of the millimeter-wave radar, and substitute them into the formula Δf = f2 - f1 to obtain the signal frequency difference Δf between the received signal frequency and the transmitted signal frequency of the millimeter-wave radar;

[0069] Calculate the flight time tf of the millimeter-wave radar signal in space, and the calculation model is specifically: where k represents the frequency modulation slope;

[0070] Based on the signal frequency difference Δf between the received signal frequency and the transmitted signal frequency of the millimeter-wave radar, and the flight time tf of the millimeter-wave radar signal in space, substitute them into the formula to obtain the distance D between the target human body and the millimeter-wave radar, where Sl represents the speed of light;

[0071] Obtain the distance d between the millimeter-wave radar antennas, the phase difference ω between the millimeter-wave radar antennas, and the wavelength λ, and substitute them into the formula to obtain the azimuth angle θ of the target human body;

[0072] Obtain the sweep period Tc of the millimeter-wave radar, and substitute it into the formula to obtain the speed V of the target human body.

[0073] S4: Judging the comprehensive coefficient of the target human body's dynamic characteristics: Based on the comprehensive coefficient of the target human body's dynamic characteristics and a preset threshold for judgment. If it is judged that the target human body is in a static state, then further execute S5; otherwise, further execute S6;

[0074] In this embodiment, it should be specifically noted that the specific content of judging the comprehensive coefficient of the target human body's dynamic characteristics is as follows:

[0075] Extract the comprehensive coefficient of the target human body's dynamic characteristics and compare it with the preset threshold. If the comprehensive coefficient of the target human body's dynamic characteristics is less than the preset threshold, then judge that the target human body is in a static state and further execute S5. If the comprehensive coefficient of the target human body's dynamic characteristics is greater than the preset threshold, then judge that the target human body is in a moving state and further execute S6.

[0076] S5: Output of the recognition result of the static human body posture: Scan the target human body in a static state, construct the posture map of the key nodes of the human body, compare it with the posture maps in the human body posture information database, and screen out the posture maps that conform to the static human body posture, and output the recognition result of the static human body posture;

[0077] In this embodiment, it should be specifically noted that the specific content of outputting the recognition result of the static human body posture is as follows:

[0078] Based on the judgment result of S4, use an infrared sensor to comprehensively scan the target human body in a static state, and extract the key nodes of the human body from the scan data; these nodes usually include joint points (such as shoulder joints, elbow joints, hip joints, knee joints, etc.) and feature points such as the head, neck, shoulders, elbows, wrists, waist, knees, and ankles;

[0079] A three-dimensional space coordinate system is established, which is represented by three axes XYZ in a rectangular coordinate system. Among them, the X-axis is in the vertical direction, that is, the standing direction of the target human body, the Y-axis is in the front-back direction of the target human body, and the Z-axis is in the left-right direction of the target human body. The key nodes of the human body extracted from the scan data are projected into the three-dimensional space coordinate system to obtain the coordinate values of the key nodes. Based on the coordinate values, the connection relationship and relative position between the key nodes are depicted, and the posture map of the target human body is constructed;

[0080] The constructed posture map of the target human body is compared with the maps in the pre-constructed human body posture information database. In the comparison results, the target maps that match the static state human body posture maps in the database are screened out, and the similarity between the posture map of the target human body and the static state map in the database is compared with a preset similarity threshold. When the similarity between the posture map of the target human body and the static state map in the database is greater than the preset similarity threshold, it is determined that the target map conforms to the static state, and the static state human body posture recognition result is output.

[0081] S6: Output of the moving state human body posture recognition result: Detect the acceleration value, center of gravity offset rate, and intensity change rate of the echo signal when the target human body is moving, analyze to obtain the motion trend evaluation coefficient SEC of the target human body, and compare it with the preset motion posture threshold interval to analyze the human body posture, and output the moving state human body posture recognition result.

[0082] In this embodiment, it should be specifically noted that the calculation steps of the motion trend evaluation coefficient of the target human body are specifically as follows:

[0083] S61: Calculate the acceleration value when the target human body is moving. Taking the ground as the reference, a space coordinate system 0XYZ is established, and the coordinate system of the target human body during movement is OA X A Y A Z , through rotating β degrees around the X-axis, rotating γ degrees around the Y-axis, and rotating α around the Z-axis for coordinate system conversion, the coordinate calculation formula is as follows:

[0084]

[0085] , where, A X 、A Y 、A Z respectively represent the acceleration values of the target human body on the X, Y, and Z axes during movement;

[0086] S62: Obtain the initial coordinates of the human body center of gravity as (x0, y0, z0). After time t′, the center of gravity coordinates become (x1, y1, z1), and substitute them into the formula to obtain the center of gravity offset rate Gsr of the target human body, where L is the height of the target human body;

[0087] S63: Obtain the echo signal strength S(t) received at time t. After the time interval [t0, t1], the echo signal strength changes from S(t0) to S(t1), and substitute it into the formula to obtain the change rate Sr of the echo signal strength of the target human body;

[0088] S63: Calculate the motion posture evaluation coefficient SEC of the target human body. The calculation model is specifically as follows:

[0089] where e represents the natural constant, and μ represents the angle between the radar beam and the human body motion direction;

[0090] In this embodiment, it should be specifically noted that the specific content output by the motion state human body posture recognition is:

[0091] Extract the motion posture evaluation coefficient of the target human body, and use the clustering analysis method to compare the motion posture evaluation coefficient of the target human body with the preset motion posture threshold interval one by one, and output the human body posture recognition results corresponding to the preset motion posture threshold intervals respectively.

[0092] Embodiment 2

[0093] Please refer to Figure 2 as shown. The present invention provides a human body posture recognition system based on millimeter-wave radar detection, including:

[0094] Build a human body posture information database module: Obtain various action posture maps of the human body in a static or moving state, and form a human body posture information database;

[0095] Identify the target human body information module: Transmit an electromagnetic wave signal within the target detection range through a millimeter-wave radar and receive the echo signal, and preprocess the echo signal, and identify and screen out the target human body information based on the preprocessed echo signal;

[0096] Target human body dynamic feature analysis module: Detect the displacement change amount, azimuth angle deviation degree and speed change rate of the target human body, and analyze and obtain the dynamic feature comprehensive coefficient DCC of the target human body based on the millimeter-wave radar echo signal parameters;

[0097] Target human body dynamic feature comprehensive coefficient judgment module: Judge based on the target human body dynamic feature comprehensive coefficient and the preset threshold. If it is judged that the target human body is in a static state, further execute the static state human body posture recognition result output module, otherwise further execute the moving state human body posture recognition result output module;

[0098] Static human body posture recognition result output module: Scan the target human body in a static state, construct a posture map of the key nodes of the human body, compare it with the posture maps in the human body posture information database, screen out the maps that conform to the static human body posture, and output the recognition result of the static human body posture;

[0099] Moving human body posture recognition result output module: Detect the acceleration value, center of gravity offset rate, and intensity change rate of the echo signal when the target human body is moving, analyze and obtain the motion trend evaluation coefficient SEC of the target human body, compare it with the preset motion posture threshold range to analyze the human body posture, and output the recognition result of the moving human body posture.

[0100] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0101] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, and all should be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A human body posture recognition method based on millimeter-wave radar detection, characterized in that, It includes the following steps: S1: Construct a human body posture information database: Obtain various action posture maps of the human body in a static or moving state, and form a human body posture information database; S2: Identify target human body information: Transmit electromagnetic wave signals and receive echo signals within the target detection range through a millimeter-wave radar, preprocess the echo signals, and identify and screen out target human body information based on the preprocessed echo signals; S3: Analysis of the dynamic characteristics of the target human body: Detect the displacement change amount, azimuth angle deviation degree, and speed change rate of the target human body, and analyze and obtain the dynamic characteristic comprehensive coefficient DCC of the target human body based on the millimeter-wave radar echo signal parameters; S4: Judgment of the dynamic characteristic comprehensive coefficient of the target human body: Based on the judgment of the dynamic characteristic comprehensive coefficient of the target human body and a preset threshold, if it is judged that the target human body is in a static state, further execute S5, otherwise further execute S6; S5: Output of the recognition result of the static state human body posture: Scan the target human body in a static state, construct a posture map of the key nodes of the human body, compare it with the posture maps in the human body posture information database, screen out the posture maps that conform to the static state human body posture, and output the recognition result of the static state human body posture; S6: Output of the recognition result of the moving state human body posture: Detect the acceleration value, center of gravity offset rate, and intensity change rate of the echo signal when the target human body is moving, analyze and obtain the motion trend evaluation coefficient SEC of the target human body, and compare it with the preset motion posture threshold interval to analyze the human body posture, and output the recognition result of the moving state human body posture.

2. The human body posture recognition method based on millimeter wave radar detection according to claim 1, wherein: The specific target detection range is the living space of the target human body. The living space of the target human body is determined as the target detection range, and the human body posture is recognized within the target detection range through a millimeter-wave radar.

3. The human body posture recognition method based on millimeter-wave radar detection according to claim 1, characterized in that: The specific content of the analysis of the dynamic characteristics of the target human body is: S31: Obtain the first distance D1 between the target human body and the millimeter-wave radar and the second distance D2 after a time interval Δt, and substitute them into the formula to obtain the displacement change ΔD of the target human body; Obtain the azimuth angle θ of the target human body and substitute it into the formula to obtain the azimuth angle offset Ao of the target human body, where θ0 represents the initial azimuth angle of the target human body; Obtain the first velocity V1 of the target human body and the second velocity V2 after a time interval Δt, and substitute them into the formula to obtain the rate of change of velocity ΔV of the target human body; S32: Obtain the millimeter-wave radar echo signal parameters, where the echo signal parameters include the intensity St, the fundamental frequency f, and the phase of the millimeter-wave radar echo signal S33: Extract the displacement change amount ΔD of the target human body, the azimuth deviation degree Ao of the target human body, the speed change rate ΔV of the target human body, the intensity St of the echo signal, the basic frequency f, and the phase Substitute them into the formula respectively to obtain the comprehensive dynamic characteristic coefficient DCC of the target human body, where e is the natural constant and α is the error factor.

4. A human body posture recognition method based on millimeter-wave radar detection according to claim 1, characterized in that: The specific content of the judgment of the dynamic characteristic comprehensive coefficient of the target human body is: Extract the dynamic characteristic comprehensive coefficient of the target human body and compare it with the preset threshold. If the dynamic characteristic comprehensive coefficient of the target human body is less than the preset threshold, it is judged that the target human body is in a static state, and further execute S5. If the dynamic characteristic comprehensive coefficient of the target human body is greater than the preset threshold, it is judged that the target human body is in a moving state, and further execute S6.

5. A human body posture recognition method based on millimeter-wave radar detection according to claim 1, characterized in that: The specific content of the output of the recognition result of the static state human body posture is: Based on the judgment result of S4, use an infrared sensor to comprehensively scan the target human body in a static state, and extract the key nodes of the human body from the scan data; Establish a three-dimensional space coordinate system, which is represented by three axes XYZ in a rectangular coordinate system, where the X-axis is the vertical direction, that is, the standing direction of the target human body, the Y-axis is the front-back direction of the target human body, and the Z-axis is the left-right direction of the target human body. Project the key nodes of the human body extracted from the scan data into the three-dimensional space coordinate system, obtain the coordinate values of the key nodes, and depict the connection relationship and relative position between the key nodes based on the coordinate values to construct the posture map of the target human body; Compare the constructed target human body posture atlas with the atlases in the pre-constructed human body posture information database. In the comparison results, screen out the target atlases that match the static state human body posture atlases in the database, and compare the similarity between the target human body posture atlas and the static state atlas in the database with a preset similarity threshold. When the similarity between the target human body posture atlas and the static state atlas in the database is greater than the preset similarity threshold, it is determined that the target atlas conforms to the static state, and the static state human body posture recognition result is output.

6. A human body posture recognition method based on millimeter-wave radar detection according to claim 1, characterized in that: The specific calculation steps of the motion trend evaluation coefficient of the target human body are as follows: S61: Calculate the acceleration value of the target human body during movement. Taking the ground as the reference, establish a spatial coordinate system 0XYZ, and the coordinate system of the target human body during movement is OA X A Y A Z , perform coordinate system conversion by rotating β degrees around the X-axis, γ degrees around the Y-axis, and α around the Z-axis. The coordinate calculation formula is as follows: Among them, A X , A Y , A Z respectively represent the acceleration values of the target human body during movement on the X, Y, and Z axes; S62: The initial coordinates of the human body's center of gravity are obtained as (x0, y0, z0). After a time t′, the center of gravity coordinates become (x1, y1, z1), and substitute them into the formula to obtain the center of gravity offset rate Gsr of the target human body, where L is the height of the target human body; S63: Obtain the echo signal strength S(t) received at time t. After the time interval [t0, t1], the echo signal strength changes from S(t0) to S(t1), and substitute it into the formula to obtain the change rate Sr of the echo signal strength of the target human body; S63: Calculate the motion trend evaluation coefficient SEC of the target human body, and the calculation model is specifically: Where e represents the natural constant and μ represents the angle between the radar beam and the human motion direction.

7. A human body posture recognition method based on millimeter-wave radar detection according to claim 1, characterized in that: The specific content of the output of the motion state human body posture recognition result is: Extract the motion trend evaluation coefficient of the target human body, and use the clustering analysis method to compare the motion trend evaluation coefficient of the target human body with the preset motion posture threshold intervals one by one, and output the human body posture recognition results corresponding to the preset motion posture threshold intervals respectively.

8. A human body posture recognition system based on millimeter-wave radar detection, which is used to implement the human body posture recognition method based on millimeter-wave radar detection described in any one of the above claims 1-7, and is characterized in that, Including: Human body posture information database construction module: Obtain various action posture atlases of the human body in a static or moving state, and form a human body posture information database; Target human body information recognition module: Transmit electromagnetic wave signals in the target detection range through a millimeter-wave radar and receive echo signals, and preprocess the echo signals, and identify and screen out target human body information based on the preprocessed echo signals; Dynamic feature analysis module of the target human body: Detect the displacement change amount, azimuth angle deviation degree and speed change rate of the target human body, and analyze and obtain the comprehensive dynamic feature coefficient DCC of the target human body based on the millimeter-wave radar echo signal parameters; Target human body dynamic feature comprehensive coefficient judgment module: Judge based on the comprehensive dynamic feature coefficient of the target human body and a preset threshold. If it is judged that the target human body is in a static state, the static state human body posture recognition result output module is further executed, otherwise the motion state human body posture recognition result output module is further executed; Static state human body posture recognition result output module: Scan the target human body in a static state, construct the posture atlas of the human body key nodes, compare it with the posture atlas in the human body posture information database and screen out the atlas that conforms to the static state human body posture, and output the static state human body posture recognition result; Motion state human body posture recognition result output module: Detect the acceleration value, center of gravity offset rate and intensity change rate of the echo signal when the target human body is moving, analyze and obtain the motion trend evaluation coefficient SEC of the target human body, and compare it with the preset motion posture threshold intervals to analyze the human body posture, and output the motion state human body posture recognition result.

Images