Motion Detection and Counting System Based on Millimeter-Wave Radar Technology

Abstract

The present invention provides a motion detection and counting system based on millimeter-wave radar technology. First, a low-power human body detection module is used to detect human activities in the target area in real time, and when human activities are detected, a trigger signal is sent to the main controller processing module to activate the radar system module. Then, the radar system module emits a frequency-modulated continuous wave signal to the target area, and the received echo signal is mixed with the transmitted signal to generate an intermediate-frequency signal. The main controller processing module performs motion posture detection and counting based on the intermediate-frequency signal to obtain the corresponding motion detection and counting results, and controls the display and storage of the motion detection and counting results. Finally, the obtained motion detection and counting results are sent to the mobile device through the wireless transceiver module. The present invention realizes the automatic detection, counting and analysis of motion actions through the ranging, speed measurement and angle measurement functions of millimeter-wave radar, improving the detection accuracy, real-time performance and user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of radar systems, and particularly to a motion detection and counting system based on millimeter-wave radar technology. Background Art

[0002] In the existing motion attitude detection and counting methods, the following technical means are usually adopted:

[0003] One is the camera monitoring system. This system relies on a high-frame-rate camera to capture the dynamic of human body jumping, and uses image processing technology to achieve the purpose of action recognition and counting. However, this technology has many disadvantages: on the one hand, it is extremely vulnerable to environmental light constraints. Once in an outdoor environment or a venue with poor lighting conditions, the clarity of the images captured by the camera will be greatly reduced, resulting in a decline in detection accuracy and a significant increase in errors. On the other hand, its demand for computing resources is quite large. The image processing process requires a large amount of computing power and poses strict requirements on the response time of the entire system, which undoubtedly increases the system operation cost and complexity. In addition, from the perspective of information acquisition, it is extremely difficult to accurately grasp the three-dimensional position and angle details of the human body with only a single camera, and it is difficult to meet the need for a comprehensive analysis of the motion attitude.

[0004] The second is the wearable sensor solution. This method uses sensors such as accelerometers and gyroscopes to collect human motion data, and then realizes the functions of jump recognition and counting through data analysis. However, this technology also has obvious shortcomings: firstly, the wearing convenience is poor. Users need to equip sensor devices in multiple parts of the body, which will interfere with their normal training rhythm to a certain extent and have an adverse impact on the sports performance. Secondly, the sensors face a high risk of failure. During the movement process, situations such as sweat erosion, violent action impact and accidental fall occur frequently, and these factors are likely to cause data loss or even physical damage to the sensors, seriously affecting the reliability of detection and counting. Summary of the Invention

[0005] In view of the above-mentioned disadvantages of the existing technology, the purpose of the present invention is to provide a motion detection and counting system based on millimeter-wave radar technology, which is used to solve the technical problems that the existing motion counting detection mainly relies on manual recording, high-frame-rate cameras or wearable devices for analysis, and there are deficiencies in real-time performance, convenience or accuracy.

[0006] To achieve the above and other related objectives, the present invention provides a motion detection and counting system based on millimeter-wave radar technology. The system includes: a low-power human detection module, a radar system module, a main controller processing module, a wireless transceiver module, and a result display module. Among them, the low-power human detection module is used to detect human activities in the target area in real time and send a trigger signal to the radar system module when human activities are detected. The radar system module is connected to the low-power human detection module and is used to transmit a frequency-modulated continuous wave signal to the target area after receiving the trigger signal, and mix the received echo signal with the transmitted signal to generate an intermediate-frequency signal. The main controller processing module is connected to the radar system module and is used to detect the target motion posture and count according to the generated intermediate-frequency signal to obtain the corresponding motion detection and counting result, and control the display and storage of the motion detection and counting result. The wireless transceiver module is connected to the main controller processing module and is used to send the obtained motion detection and counting result to the mobile device. The result display module is connected to the main controller processing module and is used to display the motion detection and counting result.

[0007] In an embodiment of the present invention, the low-power human detection module is used to detect whether there is a human body in the target area according to the real-time collected human detection data. When it is detected that there is a human body, it determines whether there is a human activity in the target area currently based on a preset human motion detection threshold, and sends a trigger signal to the radar system module when human activities are detected.

[0008] In an embodiment of the present invention, the detecting the target motion posture and counting according to the generated intermediate-frequency signal to obtain the corresponding motion detection and counting result includes: calculating the key motion information of each frame according to the generated intermediate-frequency signal; performing target motion judgment on the key motion information of each frame within a specific time period collected, and counting to obtain the corresponding motion detection and counting result.

[0009] In an embodiment of the present invention, the calculation of the key motion information for each frame based on the generated intermediate frequency signal includes: performing 1D Fourier transform and 2D Fourier transform on the intermediate frequency signal frame by frame to obtain the energy of each velocity unit corresponding to each range unit, and storing them as elements in the range-velocity unit energy matrix respectively, and each element is provided with a corresponding velocity unit subscript based on the range close to the radar subscript range and the range far from the radar subscript range; wherein, the number of elements set in the horizontal direction of the range-velocity unit energy matrix is the number of velocity units involved, and the number of elements set in the vertical direction of the matrix is the number of range units involved; determining the element with the maximum energy corresponding to each range unit in the range-velocity unit energy matrix, and recording the corresponding energy and the velocity unit subscript corresponding to the energy; counting the number of elements that meet the velocity condition far from the radar and the velocity condition close to the radar among the elements with the maximum energy corresponding to each range unit; based on the number of elements that meet the velocity condition far from the radar and the velocity condition close to the radar, calculating the number of range units with similar velocity directions and the typical similar velocity values, as well as the number of range units with the same velocity direction and the typical velocity value far from the radar; calculating the key motion information of the current frame based on the number of range units with similar velocity directions and the typical similar velocity values, as well as the number of range units with the same velocity direction and the typical velocity value far from the radar.

[0010] In an embodiment of the present invention, the determining the element with the maximum energy corresponding to each range unit in the range-velocity unit energy matrix, and recording the corresponding energy and the velocity unit subscript corresponding to the energy includes: using the maximum value calculation formula or the constant false alarm detection algorithm to calculate the maximum energy corresponding to each range unit in the range-velocity unit energy matrix and the velocity unit subscript corresponding to the energy.

[0011] In an embodiment of the present invention, the counting the number of elements that meet the velocity condition far from the radar and the velocity condition close to the radar among the elements with the maximum energy corresponding to each range unit includes: based on the velocity condition far from the radar set by the velocity tolerance threshold, the energy threshold, and the range far from the radar subscript range, determining the number of elements that meet the velocity condition far from the radar among the elements with the maximum energy corresponding to each range unit; based on the velocity condition close to the radar set by the velocity tolerance threshold, the energy threshold, and the range close to the radar subscript range, determining the number of elements that meet the velocity condition close to the radar among the elements with the maximum energy corresponding to each range unit.

[0012] In an embodiment of the present invention, calculating the key motion information of the current frame based on the number of distance units with similar velocity directions, the typical similar velocity value, the number of distance units with the same velocity direction, and the typical away velocity value includes: comparing the number of distance units with similar velocity directions and the number of distance units with the same velocity direction, taking the maximum value of the two as the first motion saliency parameter and the minimum value of the two as the second motion saliency parameter; determining whether effective motion occurs based on the first motion saliency parameter, the second motion saliency parameter, and a set threshold, and recording the corresponding ternary key motion information.

[0013] In an embodiment of the present invention, performing target motion determination on the key motion information of each frame within a specific time period collected, and obtaining the corresponding motion detection count result through counting includes: periodically intercepting the ternary key motion information of each frame within the most recent specific time period, and arranging them in reverse chronological order; using machine learning methods and / or expert system calculation methods to output the corresponding motion detection count result according to the arranged ternary key motion information of each frame.

[0014] In an embodiment of the present invention, using the expert system calculation method to output the corresponding motion detection count result according to the arranged ternary key motion information of each frame includes: determining whether there is a target motion report based on the ternary key motion information of each frame, recording the corresponding initial distance when it is determined that there is a target motion report, determining the target motion judgment parameters based on the ternary key motion information of each frame, performing target motion determination according to the effective motion conditions, and outputting the corresponding motion detection count result; where the target motion judgment parameters include: the falling start frame, the falling start distance, the rising end frame, the rising end distance, the rising start frame, and the rising start distance.

[0015] In an embodiment of the present invention, using the machine learning method to output the corresponding motion detection count result according to the arranged ternary key motion information of each frame includes: inputting the ternary key motion information of each frame into the trained machine learning model to directly output the corresponding motion detection count result; using machine learning methods and expert system calculation methods to output the corresponding motion detection count result according to the arranged ternary key motion information of each frame includes: determining whether there is a target motion report based on the ternary key motion information of each frame, recording the corresponding initial distance when it is determined that there is a target motion report, determining the target motion judgment parameters based on the ternary key motion information of each frame, and inputting the target motion judgment parameters into the trained machine learning model to output the corresponding motion detection count result.

[0016] As described above, the present invention is a motion detection and counting system based on millimeter-wave radar technology, which has the following beneficial effects: First, the low-power human detection module of the present invention detects human activities in the target area in real time, and when human activities are detected, it sends a trigger signal to the main controller processing module to activate the radar system module. Then, the radar system module emits a frequency-modulated continuous wave signal to the target area, and mixes the received echo signal with the transmitted signal to generate an intermediate frequency signal. The main controller processing module performs motion posture detection and counting based on the intermediate frequency signal to obtain the corresponding motion detection and counting results, and controls the display and storage of the motion detection and counting results. Finally, the obtained motion detection and counting results are sent to the mobile device through the wireless transceiver module. The present invention realizes the automatic detection, counting and analysis of motion actions through the ranging, speed measurement and angle measurement functions of millimeter-wave radar, improving the detection accuracy, real-time performance and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It shows a schematic flow chart of the motion detection and counting system based on millimeter-wave radar technology in an embodiment of the present invention.

[0018] Figure 2 It shows a schematic flow chart of the human detection process of the low-power human detection module in an embodiment of the present invention.

[0019] Figure 3 It shows a schematic diagram of the installation position of the radar system module in an embodiment of the present invention.

[0020] Figure 4 It shows a schematic flow chart of calculating key motion information in an embodiment of the present invention.

[0021] Figure 5 It shows a schematic flow chart of the target motion decision in an embodiment of the present invention.

[0022] Figure 6 It shows a schematic diagram of the structure of the electronic terminal in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] The following specific examples illustrate the embodiments of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0024] It should be noted that in the following description, with reference to the accompanying drawings, several embodiments of the present invention are described. It should be understood that other embodiments may also be used, and mechanical composition, structure, electrical, and operational changes may be made without departing from the spirit and scope of the present invention. The following detailed description should not be considered restrictive, and the scope of the embodiments of the present invention is only defined by the claims of the published patent. The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. Spatially related terms, such as "upper", "lower", "left", "right", "below", "beneath", "lower part", "above", "upper part", etc., may be used in the text to facilitate the description of the relationship between one element or feature shown in the figure and another element or feature.

[0025] Throughout the specification, when it is said that a certain part is "connected" to another part, this includes not only the case of "direct connection", but also the case of "indirect connection" with other elements placed in between. Additionally, when it is said that a certain part "includes" a certain constituent element, unless there is a particularly contrary record, it does not exclude other constituent elements, but means that other constituent elements may also be included.

[0026] The first, second, and third, etc. terms mentioned therein are used to illustrate various parts, components, regions, layers, and / or segments, but are not limited thereto. These terms are only used to distinguish a certain part, component, region, layer, or segment from other parts, components, regions, layers, or segments. Therefore, the first part, component, region, layer, or segment described below may refer to the second part, component, region, layer, or segment within the scope not exceeding the present invention.

[0027] Furthermore, as used herein, the singular forms "a", "an", and "the" are also intended to include the plural forms unless the context indicates otherwise. It should be further understood that the terms "comprise" and "include" indicate the presence of the stated features, operations, elements, components, items, kinds, and / or groups, but do not exclude the presence, occurrence, or addition of one or more other features, operations, elements, components, items, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or meaning any one or any combination. Thus, "A, B, or C" or "A, B, and / or C" means "any one of the following: A; B; C; A and B; A and C; B and C; A, B, and C". An exception to this definition only occurs when the combination of elements, functions, or operations is inherently mutually exclusive in some way.

[0028] The present invention provides a motion detection and counting system based on millimeter-wave radar technology. First, a low-power human body detection module is used to detect human activities in the target area in real time. When human activities are detected, a trigger signal is sent to the main controller processing module to activate the radar system module. Then, the radar system module emits a frequency-modulated continuous wave signal to the target area, and the received echo signal is mixed with the transmitted signal to generate an intermediate frequency signal. The main controller processing module performs motion posture detection and counting based on the intermediate frequency signal to obtain the corresponding motion detection and counting results, and controls the display and storage of the motion detection and counting results. Finally, the obtained motion detection and counting results are sent to the mobile device through the wireless transceiver module. The present invention realizes the automatic detection, counting and analysis of motion actions through the ranging, speed measurement and angle measurement functions of the millimeter-wave radar, improving the detection accuracy, real-time performance and user experience.

[0029] The following will be a detailed description of the embodiments of the present invention with reference to the accompanying drawings, so that those skilled in the technical field of the present invention can easily implement it. The present invention can be embodied in many different forms and is not limited to the embodiments described herein.

[0030] As Figure 1 shows a schematic structural diagram of a motion detection and counting system based on millimeter-wave radar technology in an embodiment of the present invention.

[0031] The system includes: a low-power human body detection module 1, a radar system module 2, a main controller processing module 3, a wireless transceiver module 4, and a result display module 5; wherein,

[0032] The low-power human body detection module 1 is used to detect human activities in the target area in real time and send a trigger signal to the radar system module when human activities are detected;

[0033] The radar system module 2 is used to transmit a frequency-modulated continuous wave signal to the target area after activation, and mix the received echo signal with the transmitted signal to generate an intermediate frequency signal; the radar system module 2 adopts FMCW (Frequency-Modulated Continuous-Wave) radar. FMCW radar technology has become a feasible solution for motion detection because of its accurate distance measurement, speed measurement and angle measurement capabilities, as well as strong anti-interference characteristics. FMCW radar transmits a continuous wave signal whose frequency changes linearly with time, and receives the signal reflected by the target for analysis. The basic principles of its distance measurement, speed measurement and angle measurement are as follows: The frequency of the signal emitted by the FMCW radar changes linearly with time, and the received signal will have a frequency difference with the transmitted signal due to propagation delay, namely the beat frequency. By calculating this frequency difference, combined with the swept waveform configuration and information such as the speed of light, the distance between the target and the radar can be obtained. The moving target will produce a Doppler frequency shift. By analyzing the Doppler frequency shift of the received signal, the radar can calculate the radial velocity of the target. Angle measurement principle: Calculate the azimuth or elevation angle of the target through the phase difference in the multi-antenna array.

[0034] The main controller processing module 3 is connected to the low-power human body detection module 1 and the radar system module 2, and is used to activate the radar system module 2 to work when receiving a trigger signal from the low-power human body detection module; it is also used to perform motion posture detection and counting according to the intermediate frequency signal generated by the radar system module 2 to obtain corresponding motion detection counting results, and control the display and storage of the motion detection counting results.

[0035] The wireless transceiver module 4 is responsible for data transmission and interaction with the mobile device. The wireless transceiver module 4 is connected to the main controller processing module 3, and is used to send the obtained motion detection counting results to the mobile device; specifically, the wireless transceiver module 4 may include: a WIFI transceiver module, and the motion detection counting results are sent to the mobile device for management via the WIFI module. A Bluetooth transceiver module, and the motion detection counting results are sent to the mobile terminal for management via the Bluetooth module. A command interaction module is responsible for data interaction between the WIFI / BT module and the main controller processing module 3.

[0036] The result display module 5 is connected to the main controller processing module 3 and is used to display the motion detection counting result; specifically, the result display module 5 can be a screen that displays the motion detection counting result on the screen, or it can be a voice playback device that is used to broadcast the corresponding motion detection counting result by voice.

[0037] In one embodiment, the main function of the low-power human detection module 1 is to wake up and switch the radar motion counting function through the trigger signal of the control circuit when there is human activity in the target area detected by the passive infrared sensor (PIR) or the low-power radar. The low-power human detection module 1 is used to detect whether there is a human body in the target area according to the real-time collected human detection data. When a human body is detected, it determines whether there is current human activity in the target area based on a preset human motion detection threshold, and sends a trigger signal to the main controller processing module 3 when human activity is detected.

[0038] In a preferred embodiment, the low-power human detection module 1 uses a low-power radar, and the specific detection process is as follows:

[0039] Such as Figure 2 , the low-power radar is powered on and initialized, including setting the human motion detection threshold, etc. After power-on initialization, the low-power radar is always in a working state. When the radar emits a continuous wave or a frequency-modulated continuous wave signal, if the human body is moving relative to the radar, the signal reflected from the human body will be frequency-shifted due to the Doppler effect, and the motion of the human body is detected by analyzing the changes in the signal. When the detected value exceeds the human motion detection threshold, a trigger signal is sent and the radar system module enters the motion detection and counting working mode.

[0040] In one embodiment, the radar system module 2 is set at a certain height position, and the selection of this height position is usually considered for specific detection requirements. For example, it is comprehensively determined according to factors such as the range size of the target area, the surrounding environmental conditions, and the desired detection effect. A higher position may help to expand the detection coverage range, reduce the influence of obstacles on detection, and thus be able to monitor the target area more comprehensively. For example Figure 3 The position shown.

[0041] The radar system module 2 starts when it receives the trigger signal and completes the necessary initialization operations; the radar transmitter periodically transmits / receives a frequency-modulated continuous wave swept signal, mixes the received echo signal with the transmitted signal to generate an intermediate frequency signal, and converts it into a digital signal through an analog-to-digital converter (ADC) and stores it. A typical frequency-modulated continuous wave signal transmission method is that the transmitter transmits several groups of FMCW signals according to a period (Tframe), each group of frequency-modulated continuous wave signals contains multiple chirps, and each chirp scans a certain bandwidth. A group of periodically transmitted FMCW signals is called a frame.

[0042] Preferably, after the radar system module 2 is initialized, it reads the radar calibration mark from the non-volatile memory. If the radar system module 2 has not been calibrated, it enters the radar calibration program and saves the calibration parameters to the non-volatile memory. If it has been calibrated, it directly reads the calibration value from the non-volatile memory. Calibration is performed according to the obtained calibration value, and subsequent motion attitude detection and counting are performed after calibration.

[0043] In one embodiment, the main controller processing module 3 performs motion attitude detection and counting based on the generated intermediate frequency signal to obtain corresponding motion detection counting results, including:

[0044] Step 1: Calculate the key motion information of each frame according to the generated intermediate frequency signal;

[0045] Step 2: Perform target motion judgment on the key motion information of each frame collected within a specific time period, and perform counting to obtain the corresponding motion detection counting result.

[0046] In one embodiment, as Figure 4 , Step 1 includes:

[0047] Step 11: Perform 1D Fourier transform and 2D Fourier transform on the intermediate frequency signal frame by frame to obtain the energy of each velocity unit corresponding to each range unit, and store them as elements in the range-velocity unit energy matrix respectively. And each element is set with a corresponding velocity unit subscript based on the near-radar subscript range and the far-radar subscript range;

[0048] Specifically, the space detected by the radar is divided into range units, just like concentric ring regions centered on the radar. For each such range unit, different velocity units are further subdivided to represent different velocity states that the target at that range may have. According to the basic principle of the FMCW radar receiver, perform 1D / 2D Fourier transform on the intermediate frequency signal frame by frame to obtain the energy of each velocity unit corresponding to each range unit, and store them as elements in the range-Doppler-Bin energy matrix respectively; through the above transformation, the energy magnitude of different velocity units in each range unit can be known, and this energy value reflects the possibility of the target appearing in the corresponding range and velocity states or the intensity of the target reflection signal, etc.

[0049] Among them, the energy db of each velocity unit corresponding to each range unit is stored in an N*M matrix. The number of elements M set in the vertical direction is the number of range units involved, and the number of elements N set in the horizontal direction is the number of velocity units involved, denoted as: P(m, n) where m ∈ [0, M - 1], n ∈ [0, N - 1];

[0050] And each element is provided with a corresponding velocity cell subscript based on the subscript range close to the radar and the subscript range far from the radar; In particular, for the velocity cell subscript of zero Doppler (i.e., no motion), it is denoted as n0 = N / 2. On this basis, the velocity cells far from the radar are specified as [0, n0 - 1], which can be understood as in the velocity dimension, the velocity directions corresponding to these cells are such that the target moves away from the radar; while the velocity cells close to the radar are [n0 + 1, N - 1], meaning that the velocity directions corresponding to these cells are such that the target moves closer to the radar. If in the actual signal processing or data storage process, the initial situation does not meet such requirements for setting the velocity cell subscripts, this goal can also be achieved through an operation such as Fourier transform shift (FFT shift), so as to ensure that the velocity cell subscripts in the entire range-velocity cell energy matrix are set in accordance with the established rules, facilitating subsequent accurate analysis and judgment of the target's range, velocity, motion direction, and other situations based on this information.

[0051] Step 12: Determine the element with the maximum energy corresponding to each range cell in the range-velocity cell energy matrix, and record the corresponding energy and the velocity cell subscript corresponding to this energy.

[0052] Step 13: Count the number of elements that meet the velocity conditions of moving away from the radar and the velocity conditions of moving closer to the radar among the elements with the maximum energy corresponding to each range cell.

[0053] Step 14: Based on the number of elements that meet the velocity conditions of moving away from the radar and the velocity conditions of moving closer to the radar, calculate the number of range cells with similar velocity directions and the typical similar velocity values, as well as the number of range cells with the same velocity direction and the typical velocity value of moving away.

[0054] Step 15: Calculate the key motion information of the current frame based on the number of range cells with similar velocity directions and the typical similar velocity values, as well as the number of range cells with the same velocity direction and the typical velocity value of moving away.

[0055] In one embodiment, step 12 includes: calculating the maximum energy corresponding to each range cell in the range-velocity cell energy matrix and the velocity cell subscript corresponding to this energy by using a maximum value calculation formula or a constant false alarm detection algorithm;

[0056] Specifically, for each element in P(m, n), calculate Pv(m) and Iv(m), where Pv(m) represents the maximum energy in range cell m; Iv(m) represents the velocity cell subscript corresponding to the maximum energy in range cell m.

[0057] The methods for calculating Pv(m) and Iv(m) using the maximum value calculation formula include:

[0058] P(m, n); (1)

[0059] P(m, n); (2)

[0060] In addition, the Constant False Alarm Rate (CFAR) algorithm is used to calculate Pv(m) and Iv(m). The core purpose is to detect the target signal as accurately as possible from complex background noise and interference while maintaining a constant false alarm probability. The methods include:

[0061] 1) When no valid CFAR detection points are found after CFAR detection in the given range cell m, record Pv(m) = 0 and Iv(m) = -1;

[0062] 2) When the range cell m contains a valid CFAR detection point signal, record the energy of this detection point signal as Pv(m), and mark the velocity cell index of this detection point as Iv(m);

[0063] 3) When the range cell m contains multiple valid CFAR detection point signals, regard Pv(m) and Iv(m) as queues containing multiple elements, and record the energy of each detection point signal as Pv(m) one by one, and record the corresponding velocity cell index as Iv(m) one by one.

[0064] The above CFAR detection technology can act alone in the velocity dimension or in the range-velocity dimension.

[0065] In one embodiment, step 13 includes: counting the number of elements that meet the conditions of moving away from the radar speed and moving towards the radar speed among the elements with the maximum energy corresponding to each range cell;

[0066] Based on the condition of moving away from the radar speed set by the velocity tolerance threshold, energy threshold, and the range of subscripts moving away from the radar, judge the number of elements that meet the condition of moving away from the radar speed among the elements with the maximum energy corresponding to each range cell;

[0067] Specifically, traverse all the records of Iv(m), and count the number of elements RP that simultaneously meet the following conditions of moving away from the radar speed cnt(n) :

[0068] ; (3)

[0069] ; (4)

[0070] Iv(m) ∈ [0, n0 - 1]; (5)

[0071] Wherein, in formula (3), n is the subscript of a reference speed unit for comparison, and Nthreshold is the speed tolerance threshold; P threshold is the energy threshold for admission, both are non - negative thresholds; abs is the calculation of taking the absolute value;

[0072] Based on the condition of approaching the radar set by the speed tolerance threshold, the energy threshold for admission, and the range of subscripts close to the radar, determine the number of elements that meet the speed condition of approaching the radar among the elements with the maximum energy corresponding to each range cell.

[0073] Traverse all the records of Iv(m), and count the number of elements R that simultaneously meet the following speed conditions for approaching the radar Ncnt(n) :[[]]END]]

[0074] ; (6)

[0075] ; (7)

[0076] Iv(m) ∈ [n0 + 1, N - 1]; (8)

[0077] It should be noted that if Pv(m) and Iv(m) are queues containing multiple elements, the foregoing operations act on each element of the Pv(m) and Iv(m) queues.

[0078] In one embodiment, step 14 includes:

[0079] Based on the number of elements RP cnt(n) and R Ncnt(n) that meet the conditions of moving away from the radar and approaching the radar, calculate the number of range cells P cnt,max with similar speed directions and the typical similar speed value P v,max and the number of range cells N cnt,max with the same speed direction and the typical speed value N for moving away v,max , and their calculation formulas include:

[0080] (9)

[0081] (10)

[0082] (11)

[0083] (12)

[0084] Formula (9) is to count the quantity corresponding to the case with the largest number of elements under the condition of a speed far from the radar, providing a key quantitative index for analyzing the speed distribution of the target in the direction away from the radar; formula (10) is to find the speed unit subscript when the condition of a speed far from the radar is met and the number of elements is the largest as a typical similar speed value, helping to grasp the speed characteristics of the target in this direction; formula (11) is to count the quantity corresponding to the case with the largest number of elements under the condition of a speed close to the radar, providing an important quantitative reference index for analyzing the speed distribution of the target in the direction close to the radar; formula (12) is to find the speed unit subscript when the condition of a speed close to the radar direction is met and the number of elements is the largest as a typical similar speed value, which is conducive to grasping the speed characteristics of the target in this direction.

[0085] It should be noted that if all elements in RP cnt(n) are 0, then P v,max = -1; if all elements in RN cnt(n) are 0, then N v,max = -1.

[0086] In one embodiment, step 15 includes:

[0087] Comparing the number of range units with similar speed directions and the number of range units with the same speed direction , taking the maximum value of the two as the first motion significance parameter CM and taking the minimum value of the two as the second motion significance parameter CN;

[0088] Based on the first motion significance parameter CM, the second motion significance parameter CN, and the set thresholds C threshold and D threshold to determine whether an effective motion occurs and record the corresponding three - element key motion information;

[0089] Specifically, when analyzing the motion process of a target (such as a human body) squatting or jumping, from the perspective of radar detection, there will be a phenomenon that multiple range units have similar speeds and the same speed direction. And when the motion in these main motion directions shows sufficient significance, it can be determined that there is an effective motion in the current frame of data detected by the radar. The so - called "significance" is to measure whether the characteristics of the target motion in terms of speed and range unit distribution are prominent enough through specific parameters, so as to judge whether it is a truly meaningful motion situation.

[0090] Therefore, the significance parameter is defined as:

[0091] CM = Max(P cnt,max , N cnt,max ) ; (13)

[0092] CN = Min(P cnt,max , N cnt,max ); (14)

[0093] Wherein, P cnt,max is the number of range cells with similar velocity directions obtained by previous calculation, and N cnt,max is the number of range cells with the same velocity direction.

[0094] If the significance parameter satisfies the following conditions, it is considered that there is an effective motion in this frame:

[0095] CM > Cthreshold; (15)

[0096] CM > D threshold ×CN; (16)

[0097] Wherein, C threshold and D threshold are two non - negative thresholds respectively.

[0098] If there is an effective motion in this frame, the triple - key motion information of this frame is recorded according to the following rules:

[0099] ; (17)

[0100] ; (18)

[0101] ; (19)

[0102] Wherein, CMN is regarded as a quantitative representation of relative significance; V can be understood as representing the more dominant typical velocity value in the current effective motion situation; the determination of R comprehensively considers various factors such as the energy of range cells, velocity direction, and velocity tolerance with the typical velocity value; in formula (19), the absolute value of the difference from V should be less than the velocity tolerance threshold , the maximum energy Pv(m) in range cell m should be greater than the energy threshold , and (Iv(m) - n0)×(V - n0)>0 m This condition is restricted from the perspective of velocity direction consistency to ensure that the selected range cells are consistent with the previously determined typical velocity value V in terms of velocity direction.

[0103] If there is no effective motion in this frame, the triple - key motion information of this frame is recorded according to the following rules:

[0104] CMN = 0; (20)

[0105] V = -1; (21)

[0106] R = -1; (22)

[0107] Directly assign CMN to 0, which means that when there is no effective motion in the current frame, the quantization value of its relative significance is 0. Using -1 to represent the value of V means that when there is no effective motion, there is no corresponding typical velocity value, so as to mark this state of no effective motion. Similarly, assign R to -1, indicating that in the absence of effective motion, there is no specific distance unit that meets the relevant conditions to record, so as to reflect this state of lack of relevant information of no effective motion.

[0108] In one embodiment, as Figure 5 , step 2 includes:

[0109] Step 21: Periodically intercept the three - element key motion information of each frame in the most recent specific time period, and arrange them in reverse chronological order;

[0110] Specifically, it is necessary to periodically intercept the three - element key motion information of each frame in the most recent specific time period. The specific time period can be set according to requirements, such as S seconds; the S seconds here is a set time window. By focusing on the data within this relatively short time period, the recent motion state changes of the target can be captured more timely, while avoiding excessive data volume leading to overly complex analysis. For the three - element key motion information <CMN, V, R> triple of each frame i within these S seconds, arrange them in reverse chronological order, denoted as cmn[i], v[i], and r[i], where i ∈ [0, k - 1]; among them, k represents the number of triples collected within the most recent S seconds, that is, the number of triples corresponding to the number of frames included in this time window. Arranging and storing in reverse chronological order means that the latest data will be stored at the position of i = 0, and earlier data will be stored successively backward, and the earliest data will be stored at the position of i = k - 1.

[0111] Step 22: Use machine learning methods and / or expert system calculation methods to output the corresponding counting results according to the arranged three - element key motion information of each frame.

[0112] In one embodiment, adopting the expert system calculation method to output the corresponding motion detection counting results according to the arranged three - element key motion information of each frame includes:

[0113] Based on the ternary key motion information of each frame, determine whether there is a target motion report. When it is determined that there is a target motion report, record the corresponding initial distance. Based on the ternary key motion information of each frame, determine the target motion judgment parameters, perform target motion judgment according to the effective target motion conditions, and output the corresponding motion detection count result; among them, the target motion judgment parameters include: the starting frame of falling, the starting distance of falling, the ending frame of rising, the ending distance of rising, the starting frame of rising, and the starting distance of rising.

[0114] Specifically, the target motion can be jumping or squatting. During the jumping or squatting process, during the jumping process, there is a process where the human body Doppler velocity changes from approaching (n ∈ [0, m0]) to moving away (n ∈ [m0 + 1, M - 1]), and the human body distance from the radar also has an operation of far - near - far. During the squatting process, there is a process where the human body Doppler velocity changes from moving away (n ∈ [m0 + 1, M - 1]) to approaching (n ∈ [0, m0]), and the human body distance from the radar also has an operation of near - far - near;

[0115] Now, the jumping judgment process will be described:

[0116] Based on the velocity information v(0) of the current latest frame (i.e., the frame corresponding to v(0)), preliminarily judge whether there is a jumping report. If v(0) is -1, it means that there is no typical velocity value related to effective motion, or v(0) is approaching information (v[0] ∈ [0, m0]), in this case, it is determined that there is no jumping report for this frame. Otherwise, record the initial distance at this time as R fall_end , and this distance information can be used as a starting reference value for judging the position change during the subsequent jumping process.

[0117] First, determine the starting frame of falling and the starting distance of falling. The specific method is as follows:

[0118] Start searching for v from i = 0, looking back at the data of the past M frames; here, a basic judgment condition for effective falling is set: among these M frames, if there are greater than or equal to N frames where v[i] are all in the range of moving - away information (v[i] ∈ [m0 + 1, M - 1]), and there is no frame where v[i] is in the range of approaching information (v[i] ∈ [0, m0]), then it is considered that an effective falling is recorded.

[0119] After meeting the above - mentioned preliminary judgment condition for effective falling, next, it is necessary to further accurately determine the specific frame subscript of the start of falling and the corresponding position information. Specifically: continuously search for the largest - subscript v[t] within the range of M <= t < K, and v[t] also needs to meet the following several additional limiting conditions:

[0120] 1) Among v[i] (0 <= i < t), the number of records where v[i] is -1 is less than or equal to Cv0 piece;

[0121] 2) Among v[i] (0 <= i < t), the records where v[i] is less than or equal to the information near are less than or equal to C v1 piece;

[0122] 3) v[t] itself is information far away;

[0123] And when the conditions are met, the value t is the starting frame of the fall, denoted as F dn_start， At this time, the distance record is R fn_start .

[0124] Next, determine the ending frame of the rise and the ending position of the rise. The specific method is as follows:

[0125] Start from i = F dn_start + 1 and search for the largest v[t] in the range F dn_start < t < K. And v[t] also needs to meet the following additional restrictive conditions:

[0126] 1) Among v[i] (Fdn_start < i < t), the records where v[i] is not -1 are less than or equal to Cv2 pieces;

[0127] 2) V[t] = -1;

[0128] And when the conditions are met, the value t is the ending frame of the rise, denoted as Fup_end. At this time, the distance record is Rup_end.

[0129] Next, determine the starting frame of the rise and the starting distance of the rise. The methods include:

[0130] Start searching from i = Fup_end + 1 for the largest v[t] in the range Fup_end < t < K and also need to meet the following additional restrictive conditions:

[0131] 1) Among v[i] (Fup_end < i < t), the records where v[i] is -1 are less than or equal to Cv3 pieces;

[0132] 2) Among v[i] (Fup_endt < i < t), the records where v[i] is information far away are less than or equal to Cv4 pieces;

[0133] 3) v[t] itself is information near;

[0134] And when the conditions are met, then the value t is the starting frame of the rise, denoted as F up_start , and at this time, the distance record is R up_start；

[0135] It should be noted that C herev0、 C v1、 C v2、 C v3、 C v4 is a non - negative integer and is related to the frame rate; generally speaking, the larger the frame rate, the larger the above value needs to be correspondingly.

[0136] Finally, calculate the average value Amn of cmn[i] (0 <= i <= Fup_start);

[0137] According to F fall_start、 R fall_start、 F up_end、 R up_end、 F up_start,、 R up_start、 R fall_end、 Based on the reconciliation relationship between F, R, and Amn, a series of criteria can be defined to determine whether there is a valid jump, including some or all of the following criteria:

[0138] 1) The total jump time should be greater than a given threshold: F up_start > Threshold ，1 ;

[0139] 2) The jump - up time should be greater than a given threshold: F up_start - F up_end >Threshold ,2 ;

[0140] 3) The time of staying in the air, forming zero speed, should not be too long: F up_end - F fall_start <Threshold ,3 ;

[0141] 4) The landing time should be greater than a given threshold: F fall_start >Threshold ,4 ;

[0142] 5) When staying in the air and forming zero speed, the distance should not change suddenly: abs(R up_end - R fall_start ) < Threshold ,5 ;

[0143] 6) The take - off and landing heights should not change suddenly: abs(R up_start - R fall_end ) < Threshold ,6 ;

[0144] 7) There should not be too many abnormal disturbances A mn < Threshold ,7;

[0145] If it is determined as a valid jump once, record F up_start The sampling time t corresponding to the frame where it is located. If the sampling times t recorded twice are different, it is considered a new jump; otherwise, it is considered a repeated record of a jump.

[0146] It should be noted that the squatting process is the reverse operation of the jumping operation, which means that during the squatting process, the human body will show change characteristics opposite to those of jumping in terms of speed and distance relative to the radar. During the jumping process, first approach the radar (corresponding to the speed direction being within the approaching information range), and then move away from the radar (corresponding to the speed direction being within the moving-away information range). Then for the squatting process, its reverse process is to start from the speed state of moving away from the radar first, and then change to the speed state of approaching the radar, which is derived based on the logic opposite to that of jumping in terms of the speed direction. During the squatting process, the distance between the human body and the radar will show a near-far-near change. That is, at the beginning, it is relatively close to the radar, then gradually becomes farther away, and finally returns to a relatively close position, which is exactly the opposite of the distance change during the jumping process. Therefore, based on the detailed processing methods of each stage of the jumping process above, it is possible to correspondingly infer how to determine the key stages such as the start and end of squatting according to the reverse logic of the above speed and distance changes. This will not be elaborated here. For example, during the jumping process, the start frame of the fall is determined by judging that the speed direction remains away within a certain number of frames. Then during the squatting process, the relevant frame information of the start of the squat is determined by judging that the speed direction remains close within a certain number of frames; another example is that during the jumping process, the end frame of the ascent is determined based on conditions such as the speed value being -1. During squatting, perhaps the end frame of the squat can be judged according to similar but reverse speed-related conditions. And for operations such as recording the distance at the corresponding stage, it is also possible to make reasonable settings and inferences according to the reverse situation of the distance change, so as to construct a complete processing method for the squatting process, which is used to analyze the manifestation of the human motion behavior of squatting in the radar detection data and related feature judgments, etc.

[0147] In one embodiment, when analyzing the human motion situation, in addition to the expert system calculation method based on rules and condition judgments mentioned above, a machine learning method can also be adopted. Its core idea is to directly use the ternary key motion information <CMN, V, R> of each frame arranged in sequence as input data and provide it to a trained machine learning model. The model automatically analyzes these data features and then outputs the corresponding counting result, that is, to complete the statistics of the target motion situation (such as the number of jumps, squats, etc.). In this embodiment, a classifier can be used to perform machine learning and classification on the ternary key motion information of each frame and give a criterion. The classifier can adopt a support vector machine (SVM), logistic regression, and a long short-term memory neural network (LSTM).

[0148] In one embodiment, when analyzing the target motion condition and obtaining the target motion judgment parameters, a method combining machine learning with expert system calculation is adopted. Although both are based on the foundation of machine learning technology, they are different in terms of data feature utilization. Instead of directly using the previously mentioned triple sequence <CMN, V, R> as learning features, a series of target motion judgment parameters recorded by expert system means are selected, including: F fall_start、 R fall_start、 F up_end、 R up_end、 F up_start,、 R up_start、 R fall_end、 A mn Data. These data are used as input data and supplied to the trained machine learning model. The model automatically analyzes these data features and then outputs the corresponding counting results, that is, completes the statistics of the target motion condition (such as the number of jumps, squats, etc.).

[0149] In one embodiment, after obtaining the motion detection counting result, it is judged whether it is necessary to exit the step counting mode. If not, the next frame signal is transmitted and subsequent processing is performed.

[0150] The motion detection counting system based on millimeter wave radar technology provided by the embodiments of the present invention can be implemented on the terminal side or the server side. As for the hardware structure of the electronic terminal, please refer to Figure 6 , which is an optional hardware structure schematic diagram of the motion detection counting terminal 1000 for millimeter wave radar technology provided by the embodiments of the present invention. The terminal 1000 can be a mobile phone, a computer device, a tablet device, a personal digital processing device, a factory background processing device, etc. The terminal 1000 includes: at least one processor 1001, a memory 1002, at least one network interface 10010, and a user interface 1009. Each component in the device is coupled together through a bus system 1005. It can be understood that the bus system 1005 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 1005 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clear illustration, in Figure 6 all kinds of buses are labeled as the bus system.

[0151] Among them, the user interface 1009 may include a display, a keyboard, a mouse, a trackball, a click gun, a key, a button, a touchpad, or a touch screen, etc.

[0152] It can be understood that the memory 1002 can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM, Read Only Memory), a programmable read-only memory (PROM, Programmable Read-Only Memory), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM, Static Random Access Memory), synchronous static random access memory (SSRAM, Synchronous Static Random Access Memory). The memory described in the embodiments of the present invention is intended to include but not limited to these and any other suitable categories of memory.

[0153] The memory 1002 in the embodiments of the present invention is used to store various categories of data to support the operation of the terminal 1000. Examples of these data include: any executable program for operating on the terminal 1000, such as the operating system 10021 and the application program 10022; the operating system 10021 contains various system programs, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks. The application program 10022 can include various application programs, such as a media player (Media Player), a browser (Browser), etc., for implementing various application services. The motion detection and counting system based on millimeter-wave radar technology provided by the embodiments of the present invention can be included in the application program 10022.

[0154] The method disclosed in the above embodiments of the present invention can be applied to or implemented by the processor 1001. The processor 1001 may be an integrated circuit chip with signal processing capabilities. During implementation, the steps of the above method can be completed by the integrated logic circuit in hardware or instructions in software form in the processor 1001. The above-mentioned processor 1001 can be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 1001 can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor 1001 can be a microprocessor or any conventional processor, etc. Combining the steps of the accessory optimization method provided in the embodiments of the present invention can be directly embodied as being completed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium, and this storage medium is located in the memory. The processor reads the information in the memory and combines its hardware to complete the steps of the foregoing method.

[0155] In an exemplary embodiment, the terminal 1000 can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), or complex programmable logic devices (CPLDs) for executing the foregoing method.

[0156] Those of ordinary skill in the art can understand that all or part of the steps for implementing the above method embodiments can be completed by hardware related to a computer program. The foregoing computer program can be stored in a computer-readable storage medium. When this program is executed, it executes the steps including those of the above method embodiments; and the foregoing storage medium includes various media such as ROM, RAM, magnetic disks, or optical discs that can store program code.

[0157] In the embodiments provided by this application, the computer-readable and writable storage medium may include a read-only memory, a random access memory, an EEPROM, a CD-ROM, or other optical disc storage devices, a magnetic disk storage device, or other magnetic storage devices, flash memory, a USB flash drive, a mobile hard disk, or any other medium that can be used to store desired program code in the form of instructions or data structures and can be accessed by a computer. Additionally, any connection may be appropriately referred to as a computer-readable medium. For example, if instructions are sent from a website, server, or other remote source using coaxial cables, fiber optic cables, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cables, fiber optic cables, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of the medium. However, it should be understood that computer-readable and writable storage media and data storage media do not include connections, carrier waves, signals, or other transient media, but are intended to refer to non-transient, tangible storage media. As used in the application, magnetic disks and optical discs include compact discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, where magnetic disks typically reproduce data magnetically, while optical discs reproduce data optically using lasers.

[0158] Compared with the prior art, the present invention has the following advantages:

[0159] 1. Provide a low-power consumption and millimeter-wave radar cooperation solution: With its excellent low-power consumption characteristics, the low-power radar can always remain in an operating state and continuously monitor the surrounding environment. In contrast, the millimeter-wave radar has relatively high power consumption during operation, so a selective working mechanism is adopted and it is only activated when necessary. By cleverly using the low-power radar as a trigger source to activate the millimeter-wave radar to work, the two complement each other, effectively overcoming the problem of excessive overall power consumption and providing a solid guarantee for the continuous and stable operation of the system.

[0160] 2. Non-contact precise measurement: Relying on the cutting-edge millimeter-wave radar technology, the present invention has pioneered the non-contact precise detection of the number of human movements. This innovative measure completely abandons the inconvenience and discomfort that may be brought by traditional contact measurement methods, comprehensively improves the comfort during the movement detection process, and at the same time greatly simplifies the operation process, giving users a more convenient and efficient usage experience.

[0161] 3. Efficient Data Transmission and Intelligent Management: An advanced data transmission module is integrated within the system. This module acts like a bridge, seamlessly connecting the internal detection system with various external devices. Whether it is a convenient mobile phone APP or a powerful computer terminal, the motion data can be transmitted to these external devices in real-time and stably, enabling users to comprehensively and precisely control the motion information in real-time anytime and anywhere, fully meeting diverse usage requirements.

[0162] 4. Multivariate Motion Counting Algorithm System: The present invention constructs a comprehensive and multi-level motion counting algorithm framework. Among them, the single-frame decision mechanism based on the unit maximum value statistical principle can accurately capture key information at the single-frame data level; the jump / squat decision-making strategy based on the expert system fully draws on the knowledge of domain experts to deeply analyze complex motion states; and there is also the cutting-edge technology of machine learning based on abstract features. By means of a big data training model, it can intelligently identify various motion patterns. The three work together synergistically to ensure a high degree of accuracy in motion counting.

[0163] 5. Wide Application: The present invention has excellent cross-domain adaptability and is deeply integrated into multiple key fields such as medical and health, sports, and school education. Whether it is assisting in motion monitoring during medical rehabilitation, providing accurate data feedback for sports training, or realizing the quantitative management of students' motion states in the campus education scenario, it demonstrates incomparable practical value, injecting new vitality into the development of various industries and creating broad prospects.

[0164] In summary, for the motion detection and counting system based on millimeter-wave radar technology of the present invention, first, the low-power human detection module detects the human activities in the target area in real-time, and when human activities are detected, it sends a trigger signal to the main controller processing module to activate the radar system module. Then, the radar system module emits a frequency-modulated continuous-wave signal to the target area, and mixes the received echo signal with the transmitted signal to generate an intermediate-frequency signal. The main controller processing module performs motion posture detection and counting based on the intermediate-frequency signal to obtain the corresponding motion detection and counting results, and controls the display and storage of the motion detection and counting results. Finally, the obtained motion detection and counting results are sent to the mobile device through the wireless transceiver module. The present invention realizes the automatic detection, counting, and analysis of motion actions through the ranging, speed measurement, and angle measurement functions of the millimeter-wave radar, improving the detection accuracy, real-time performance, and user experience. Therefore, the present invention effectively overcomes various shortcomings in the prior art and has high industrial utilization value.

[0165] The above embodiments are only used to exemplarily illustrate the principles and effects of the present invention, rather than to limit the present invention. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes completed by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed by the present invention should still be covered by the claims of the present invention.

Claims

1. A motion detection and counting system based on millimeter-wave radar technology, characterized in that, The system includes: a low-power human detection module, a radar system module, a main controller processing module, a wireless transceiver module, and a result display module; among them, the low-power human detection module is used to detect human activities in the target area in real time, and send a trigger signal to the radar system module when human activities are detected; The radar system module is used to transmit a frequency-modulated continuous wave signal to the target area after being activated, and mix the received echo signal with the transmitted signal to generate an intermediate frequency signal; The main controller processing module is connected to the low-power human detection module and the radar system module, and is used to activate the radar system module to work after receiving the trigger signal; it is also used to perform target motion posture detection and counting based on the generated intermediate frequency signal to obtain corresponding motion detection counting results, and control the display and storage of the motion detection counting results; The wireless transceiver module is connected to the main controller processing module and is used to send the obtained motion detection counting results to the mobile device; The result display module is connected to the main controller processing module and is used to display the motion detection counting results; Among them, the performing target motion posture detection and counting based on the generated intermediate frequency signal to obtain corresponding motion detection counting results includes: calculating the key motion information of each frame based on the generated intermediate frequency signal; performing target motion judgment on the key motion information of each frame collected within a specific time period, and performing counting to obtain the corresponding motion detection counting results; The calculating the key motion information of each frame based on the generated intermediate frequency signal includes: performing 1D Fourier transform and 2D Fourier transform on the intermediate frequency signal frame by frame to obtain the energy of each velocity unit corresponding to each range unit, and storing them as elements in the range-velocity unit energy matrix respectively, and each element is set with a corresponding velocity unit subscript based on the near-radar subscript range and the far-radar subscript range; where the number of elements set in the horizontal direction of the range-velocity unit energy matrix is the number of velocity units involved, and the number of elements set in the vertical direction of the matrix is the number of range units involved; determining the element with the maximum energy corresponding to each range unit in the range-velocity unit energy matrix, and recording the corresponding energy and the velocity unit subscript corresponding to the energy; counting the number of elements that meet the far-radar velocity condition and the near-radar velocity condition among the elements with the maximum energy corresponding to each range unit; based on the number of elements that meet the far-radar velocity condition and the near-radar condition, calculating the number of range units with similar velocity directions and the typical similar velocity value, as well as the number of range units with the same velocity direction and the typical far velocity value; calculating the key motion information of the current frame based on the number of range units with similar velocity directions and the typical similar velocity value, as well as the number of range units with the same velocity direction and the typical far velocity value.

2. The motion detection and counting system based on millimeter-wave radar technology according to claim 1, characterized in that The low-power human detection module is used to detect whether there is a human body in the target area according to the real-time collected human detection data. When it detects the presence of a human body, it determines whether there is current human activity in the target area based on a preset human motion detection threshold, and sends a trigger signal to the radar system module when it detects human activity.

3. The motion detection and counting system based on millimeter-wave radar technology according to claim 1, characterized in that Determining the element with the maximum energy corresponding to each range cell in the range-velocity unit energy matrix and recording the corresponding energy and the subscript of the velocity cell corresponding to this energy includes: Calculating the maximum energy corresponding to each range cell in the range-velocity unit energy matrix and the subscript of the velocity cell corresponding to this energy by using a maximum value calculation formula or a constant false alarm detection algorithm.

4. The motion detection and counting system based on millimeter-wave radar technology according to claim 1, characterized in that, Counting the number of elements that meet the conditions of the velocity away from the radar and the conditions of the velocity towards the radar among the elements with the maximum energy corresponding to each range cell includes: Based on the velocity condition away from the radar set by the velocity tolerance threshold, the energy threshold, and the range of subscripts away from the radar, determining the number of elements that meet the velocity condition away from the radar among the elements with the maximum energy corresponding to each range cell; Based on the condition towards the radar set by the velocity tolerance threshold, the energy threshold, and the range of subscripts towards the radar, determining the number of elements that meet the velocity condition towards the radar among the elements with the maximum energy corresponding to each range cell.

5. The motion detection and counting system based on millimeter-wave radar technology according to claim 1, characterized in that Calculating the key motion information of the current frame based on the number of range cells with similar velocity directions, the typical similar velocity values, the number of range cells with the same velocity direction, and the typical velocity values away from the radar includes: Comparing the number of range cells with similar velocity directions and the number of range cells with the same velocity direction, taking the maximum value of the two as the first motion significance parameter and the minimum value of the two as the second motion significance parameter; Based on the first motion significance parameter, the second motion significance parameter, and the set threshold, determining whether there is effective motion and recording the corresponding three-element key motion information.

6. The motion detection and counting system based on millimeter-wave radar technology according to claim 5, characterized in that, Performing target motion judgment on the key motion information of each frame within a specific time period collected, and performing counting to obtain the corresponding motion detection counting result includes: Periodically intercepting the three-element key motion information of each frame within the most recent specific time period and arranging them in reverse chronological order; Using a machine learning method and / or an expert system calculation method to output the corresponding motion detection counting result according to the arranged three-element key motion information of each frame.

7. The motion detection and counting system based on millimeter-wave radar technology according to claim 6, characterized in that, Using an expert system calculation method to output the corresponding motion detection counting result according to the arranged three-element key motion information of each frame includes: Based on the three-element key motion information of each frame, determining whether there is a target motion report. When it determines that there is a target motion report, recording the corresponding initial distance, and determining the target motion judgment parameters based on the three-element key motion information of each frame, and performing target motion judgment according to the effective motion condition, and outputting the corresponding motion detection counting result; where the target motion judgment parameters include: the starting frame of the fall, the starting distance of the fall, the ending frame of the rise, the ending distance of the rise, the starting frame of the rise, the starting distance of the rise.

8. The motion detection and counting system based on millimeter-wave radar technology according to claim 7, characterized in that, The method of using machine learning to output the corresponding motion detection count result according to the three - element key motion information of each arranged frame includes: inputting the three - element key motion information of each frame into the trained machine learning model and directly outputting the corresponding motion detection count result; The method of using machine learning and expert system calculation methods to output the corresponding motion detection count result according to the three - element key motion information of each arranged frame includes: judging whether there is a target motion report based on the three - element key motion information of each frame, recording the corresponding initial distance when it is judged that there is a target motion report, determining the target motion judgment parameter based on the three - element key motion information of each frame, and inputting the target motion judgment parameter into the trained machine learning model to output the corresponding motion detection count result.

Images