Target identification method, device, electronic device and storage medium

Abstract

The present application relates to the field of radar signal processing technology, and in particular to a target identification method, device, electronic device and storage medium. The reflection point data of the detection object is extracted through a sliding window, and at least one candidate valid target point set is determined based on the reflection point data in the sliding window. The target corresponding to the candidate valid target point set is a candidate valid target. Since the sliding window is a time window for sampling reflection point data, and one time window corresponds to a time period; therefore, the embodiment of the present application can obtain the reflection point data that appears within a certain time period, and determine the candidate valid target point set based on the reflection point data that appears within a period of time. Compared with the method of determining the valid target by the reflection point data that appears at a certain moment, the embodiment of the present application determines the candidate valid target by the reflection point data within a period of time, and the candidate valid target identified by the method of determining the candidate valid target by the reflection point data within a period of time has higher accuracy.

Description

Technical Field

[0001] The present invention relates to the field of radar signal processing technology, and in particular to a target recognition method, device, electronic equipment and storage medium. Background Art

[0002] Millimeter waves are electromagnetic waves between infrared light and microwave frequencies. Their wavelengths range from 1 to 10 mm, corresponding to a frequency range of 30-300 GHz. Millimeter-wave radar is an electronic device that uses millimeter waves to detect targets. When a millimeter-wave radar is operating, the radar antenna radiates millimeter waves into the atmosphere in a directionally directed manner. Millimeter waves propagate at nearly the speed of light. If a target happens to be within the millimeter-wave beam emitted by the radar antenna, it intercepts a portion of the millimeter wave and scatters it in all directions, forming a reflected signal. The radar antenna then processes the received reflected signal to obtain information about the detected object.

[0003] In related technologies, millimeter-wave radiation emitted by radar antennas can be reflected by objects other than the target, causing the reflected millimeter-wave signals to propagate along multiple paths. When these reflected signals, transmitted along different paths, reach the millimeter-wave radar's receiving end, they overlap and interfere with each other based on their respective phases, a phenomenon known as multipath interference. Multipath interference causes the amplitude and phase of the target's reflected signal to vary when the millimeter-wave radar detects a target, leading to detection errors. Summary of the Invention

[0004] To overcome the detection errors caused by multipath interference when millimeter-wave radar detects targets, embodiments of the present invention provide a target recognition method, device, electronic device, and storage medium. These methods can extract reflection point data appearing within a time period through a sliding window, and determine a set of candidate valid target points based on the reflection point data appearing within a time period, thereby improving the accuracy of target recognition.

[0005] In order to solve the above technical problems, the embodiments of the present invention provide the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a method for identifying a target, applied to an electronic device, the method comprising:

[0007] Extracting reflection point data of the detection object through a sliding window, wherein the sliding window is a time window for sampling the reflection point data;

[0008] At least one candidate valid target point set is determined according to the reflection point data in the sliding window, and the targets corresponding to the candidate valid target point set are candidate valid targets.

[0009] In a second aspect, an embodiment of the present invention provides a target identification device, applied to a millielectronic device, comprising:

[0010] an extraction module, the extraction module being configured to extract the reflection point data according to a sliding window, the sliding window being a time window for sampling the reflection point data;

[0011] A determination module is configured to determine at least one candidate valid target point set based on the reflection point data within the sliding window, wherein the target corresponding to the candidate valid target point set is a candidate valid target.

[0012] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the target identification method as described in the first aspect of the present invention.

[0013] In a fourth aspect, an embodiment of the present invention further provides a non-volatile computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are executed, the target identification method as described in the first aspect of the present invention can be executed.

[0014] The beneficial effects of the embodiments of the present application are as follows: Different from the prior art, the embodiments of the present application provide a target identification method, device, electronic device and storage medium, which can extract the reflection point data of the detection object through a sliding window, and determine at least one candidate valid target point set based on the reflection point data in the sliding window, and the target corresponding to the candidate valid target point set is the candidate valid target. Since the sliding window is a time window for sampling reflection point data, and a time window corresponds to a time period; therefore, the embodiments of the present application can obtain the reflection point data that appears within a certain time period, and determine the candidate valid target point set based on the reflection point data that appears within a period of time. Compared with the method of determining the valid target by the reflection point data that appears at a certain moment, the accuracy of the candidate valid targets identified by the embodiment of the present application by the reflection point data within a period of time is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0016] Figure 1This is an application scenario provided by an embodiment of the present application;

[0017] Figure 2 This is the hardware structure of a millimeter wave radar device provided by an embodiment of the present application;

[0018] Figure 3a This is a flowchart of a target identification method provided by an embodiment of the present application;

[0019] Figure 3b is a flowchart of a target identification method provided by another embodiment of the present application;

[0020] Figure 4 1 is a flow chart of a method for determining whether a reflection point in a target cluster is a candidate valid target point set, provided by an embodiment of the present application;

[0021] Figure 5 This is a schematic diagram of a sliding window provided by an embodiment of the present application;

[0022] Figure 6 is a schematic structural diagram of a target recognition device provided by an embodiment of the present application;

[0023] Figure 7 is a schematic structural diagram of a target recognition device provided by another embodiment of the present application;

[0024] Figure 8 This is the hardware structure of an electronic device for executing a target identification method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0026] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as those commonly understood by those skilled in the art to which the present invention pertains. The terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit the present invention.

[0027] In describing the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the technical features being referred to. Furthermore, terms such as "horizontal" and "vertical" indicating orientations or positional relationships are based on the orientations and positional relationships shown in the accompanying drawings and are intended solely for the purpose of describing the present invention or facilitating description. They do not indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limiting the present invention.

[0028] In the embodiments of the present application, the electronic device for executing the target identification method exists in various forms, including but not limited to mobile communication devices, servers, or other electronic devices with computing and processing functions. Specifically, the electronic device may be a radar.

[0029] The wave radar involved in the embodiments of the present application can also be called a detector or a detection device. Its working principle is to transmit radar signals through a radar antenna and detect targets in the detection object by receiving reflected signals, wherein the reflected signal is the electromagnetic wave after the radar signal is reflected by the detection object. Radar signals can also be called radio signals or electromagnetic wave signals; for the convenience of explanation, the embodiments of the present application are uniformly referred to as radar signals. The emission of radar signals is periodic. For example, for a sawtooth wave radar signal, the duration of a complete sawtooth waveform can be understood as the emission period of the radar signal; for another example, for a triangular wave radar signal, the duration of a complete triangular waveform can be understood as the emission period of the triangular wave radar signal.

[0030] The radar involved in the embodiments of the present application may be a millimeter wave radar. Millimeter waves have the following characteristics: 1) Large bandwidth: rich frequency domain resources and low antenna side lobes, which are conducive to imaging or quasi-imaging; 2) Short wavelength: the volume of the radar equipment and the antenna aperture can be reduced, and the weight can be reduced; 3) Narrow beam: Under the same antenna size, the beam of the millimeter wave is much narrower than that of the micron wave, the radar resolution is high, and the measurement accuracy can reach the millimeter level; 4) Strong penetration: Compared with laser radar and optical systems, it has better ability to penetrate smoke, dust and clothing, and can work all day and night. The characteristics related to millimeter waves are very suitable for indoor human body detection, such as human positioning, tracking or vital sign detection.

[0031] Millimeter-wave radars produce multipath effects during target detection, leading to radar signal fading and phase shifts. Multipath occurs when electromagnetic waves propagate along different paths, with each component arriving at the receiver at different times. These components, due to their phases, interfere with each other, distorting or inaccurate signals received by the radar. Consequently, multipath makes it difficult for radars to accurately determine the location of the reflection point corresponding to the reflected signal, thus affecting the efficiency and accuracy of object recognition. Indoor environments are relatively complex, with millimeter waves potentially reflecting off walls, floors, ceilings, and various furniture items. Therefore, multipath interference is particularly pronounced when millimeter-wave radars are used to detect objects indoors.

[0032] See also Figure 1 , Figure 1 The application scenarios of the embodiments of the present application are schematically shown, such as Figure 1 As shown, the millimeter wave radar 10 is set indoors. The millimeter wave radar 10 can be an independent detection device or integrated into other products. For example, the millimeter wave radar 10 can be integrated into home appliances, such as smart robots. The target 20 to be detected is a human body or an object. Figure 1 The target 20 is a human body as an example for explanation. Figure 1 As shown, the millimeter-wave radar 10 can transmit radar signals to the detection area where the target 20 is located and receive reflected signals of the radar signals. The reflected signal from the target 20 can reach the millimeter-wave radar 10 directly along path a, or can reach the radar after being reflected from the wall 30 along path b, or can reach the millimeter-wave radar 10 after being reflected from a painting 40 mounted on the wall 30 along path c. The reflected signals received by the millimeter-wave radar 10 include reflected signals returned along paths a, b, and c. Among them, the reflected signals reaching the millimeter-wave radar 10 along paths b and c are interference signals caused by the multipath effect. If the reflected signal containing interference signals is directly used to detect the target 20, the accuracy and efficiency of the millimeter-wave radar 10 in identifying the target 20 will be affected.

[0033] In order to reduce the impact of multipath effects on the accuracy and efficiency of target recognition, an embodiment of the present application provides a target recognition method that can obtain reflection points based on the reflection signal of a radar signal, extract the reflection points according to a sliding window, and determine a candidate valid target point set from the reflection points, thereby filtering out false points other than the candidate valid target point set from the reflection points; then, based on the number of sliding windows that extract the same candidate valid target point set, a second valid point in the candidate valid target point set is selected to filter out false points in the candidate valid target point set; wherein the detection object corresponding to the second valid point is a valid target. To facilitate the reader's understanding of the present invention, it is described below in conjunction with specific embodiments.

[0034] Figure 2 The hardware structure of the millimeter-wave radar is schematically shown. Figure 2 The millimeter wave radar 200 can generate a radar signal and transmit the radar signal to the area that the millimeter wave radar 200 is monitoring. According to the waveform, the radar signal emitted by the millimeter wave radar 200 is divided into a pulse wave (chirp) signal and a frequency modulated continuous wave (CW) signal. If the radar signal is a pulse wave, then when the target is very close, the time difference between the transmitted pulse wave and the received pulse wave is very small, requiring the radar to adopt high-speed signal processing technology. The structure of the short-range pulse radar becomes very complex and the cost also increases significantly. Therefore, the millimeter wave radar usually adopts a frequency modulated continuous wave, which has a simple structure, low cost, and is suitable for close-range detection. The embodiment of the present application is described by taking the radar signal as a frequency modulated continuous wave as an example.

[0035] like Figure 2 As shown, the millimeter wave radar 200 includes: an oscillator 201, a directional coupler 202, a transmitting antenna 203, a receiving antenna 204, a mixer 205, a processor 206 and a memory 207. Those skilled in the art will understand that Figure 2 The structure shown does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown in the figure, or some components may be combined or separated, or the components may be arranged differently.

[0036] The oscillator 201 is used to generate a frequency modulated continuous wave (FMCW) and output the FMCW to the directional coupler 202. In some embodiments, the FMCW can be a linear FMCW. The frequency of the linear FMCW increases or decreases linearly with time, that is, within a unit of time, the frequency of the FMCW is linearly related to time, or the absolute value of the frequency change of the FMCW is the same. The waveform of the FMCW is generally a sawtooth wave or a triangle wave; for a sawtooth wave, the frequency change is the same; for a triangle wave, the absolute value of the frequency change is the same on the rising edge and the falling edge.

[0037] The directional coupler 202 outputs the FM continuous wave to the mixer 205 as a local oscillator signal and outputs the FM continuous wave to the transmitting antenna 203 for transmission. The FM continuous wave is reflected by the detection object and becomes a reflected wave of the FM continuous wave.

[0038] The receiving antenna 204 receives the reflected wave of the FMCW and outputs it to the mixer 205 .

[0039] The mixer 205 mixes the FM continuous wave with the reflected wave of the FM continuous wave to obtain an intermediate frequency (IF) signal, wherein the IF signal is a signal formed by the frequency difference between the FM continuous wave and the reflected wave of the FM continuous wave at the same time. That is, the frequency of the IF signal is the frequency difference between the FM continuous wave and the reflected wave of the FM continuous wave at the same time.

[0040] The mixer 205 filters the intermediate frequency signal obtained by mixing through a low-pass filter, and then outputs it to the processor 206 after amplification. After the processor 206 processes the intermediate frequency signal (for example, performing fast Fourier transform and spectrum analysis on the intermediate frequency signal), it obtains at least one of the distance information, speed information and angle information of the detection object.

[0041] In the embodiments of the present application, the distance information, speed information and angle information of the detection object, or the distance, speed and angle of the detection object, are all relative to the millimeter-wave radar that transmits frequency-modulated continuous waves, and the present application does not limit the specific names.

[0042] The following takes the triangular wave waveform of the frequency modulated continuous wave as an example to introduce the working principle of the millimeter wave radar in detail.

[0043] The period of the frequency modulated continuous wave (triangle wave) is T. The frequency of the frequency modulated continuous wave increases linearly by ΔF as time increases within the time unit [0, T / 2]. The frequency of the frequency modulated continuous wave decreases linearly by ΔF as time increases within the time unit [T / 2, T]. Here, ΔF is the maximum variation range of the frequency of the frequency modulated continuous wave.

[0044] If the detection object is relatively stationary with respect to the millimeter-wave radar, the shape of the FMCW and the reflected wave of the FMCW are the same, but there is a time delay Δt in time.

[0045] Frequency modulated continuous wave x1 is:

[0046]

[0047] The reflected wave x2 of the FMCW is:

[0048]

[0049] Where ω1(t) is the angular velocity of the FMCW x1, is the initial phase of the frequency modulated continuous wave x1; ω2(t) is the angular velocity of the frequency modulated continuous wave x2, is the initial phase of the frequency modulated continuous wave x2.

[0050] The time delay Δt between the FMCW x1 and its reflected wave x2 and the distance R of the detection object satisfy:

[0051] Δt=2R / c (3)

[0052] Where c is the speed of light.

[0053] The FM continuous wave x1 and the reflected wave x2 of the FM continuous wave are mixed in the mixer to obtain the intermediate frequency signal x out for:

[0054]

[0055] The frequency IF of the intermediate frequency signal is the product of the slope s of the frequency of the frequency modulated continuous wave and the time delay Δt, that is:

[0056]

[0057] Wherein, T is the period of the triangle wave, and ΔF is the maximum variation range of the FMCW.

[0058] Therefore, the distance R of the detected object is:

[0059]

[0060] The above derivation shows that for a detection object that is relatively stationary relative to the millimeter-wave radar, the frequency difference between the FM continuous wave and the reflected wave of the FM continuous wave at the same time (the frequency of the intermediate frequency signal IF) and the time delay Δt between the FM continuous wave and the reflected wave of the FM continuous wave are linearly related. That is, the farther the distance to the target object, the later the reflected wave of the FM continuous wave is received by the millimeter-wave radar, and the greater the frequency difference between the FM continuous wave and the reflected wave of the FM continuous wave at the same time (the frequency of the intermediate frequency signal IF). Therefore, the distance to the detection object can be obtained by determining the magnitude of the intermediate frequency signal frequency IF.

[0061] If the detection object is in relative motion with respect to the radar, the frequency of the reflected wave of the FMCW includes the Sopler frequency shift f caused by the relative motion of the target object. d .

[0062] Therefore, the frequency f of the intermediate frequency signal corresponding to the rising edge of the triangle wave is b+ for:

[0063] f b+ =IF-f d (7)

[0064] The frequency f of the intermediate frequency signal corresponding to the falling edge of the triangle wave b- for:

[0065] f b- =IF+f d (8)

[0066] Where IF is the frequency of the intermediate frequency signal when the target object is relatively stationary relative to the radar, f d =2fv / c is the Doppler shift, and its sign is related to the direction of the target object's relative motion with respect to the radar, where f is the center frequency of the FMCW and v is the velocity of the detected object.

[0067] From the above formulas (7), (8) and (5), we can get the distance R of the detection object in relative motion relative to the radar:

[0068]

[0069] From the above formulas (7) and (8) and the Doppler frequency shift, we can also get the speed of the detection object that can move relative to the radar:

[0070]

[0071] It can be seen from the above formulas (9) and (10) that, for a detection object that is in relative motion with respect to the radar, the distance and speed of the detection object relative to the radar can be obtained by detecting the frequency of the intermediate frequency signal at the rising edge and the frequency of the intermediate frequency signal at the falling edge of the triangular wave.

[0072] The following uses the sawtooth frequency modulated continuous wave as an example to introduce the working principle of millimeter wave radar.

[0073] For sawtooth waves, the distance measurement principle is similar to that of triangle waves. Let the time delay between the FM continuous wave and the reflected wave of the FM continuous wave be τ, and the period of the sawtooth wave be T c .

[0074] Frequency modulated continuous wave x1 is:

[0075]

[0076] The reflected wave x2 of the FMCW is:

[0077]

[0078] Where ω1 is the angular velocity of the FMCW x1, is the initial phase of the frequency modulated continuous wave x1.

[0079] Initial phase of the IF signal is the phase difference between the FM continuous wave x1 and the FM continuous wave x2:

[0080]

[0081] Since τ = 2R / c, the initial phase of the intermediate frequency signal can be further obtained

[0082]

[0083] The distance of the target object is:

[0084]

[0085] Where λ = c / f c is the wavelength of the FMCW, R is the distance to the target object, f c is the center frequency of the FMCW.

[0086] From the above derivation, it can be seen that the relative distance between the target object and the radar can be obtained by detecting the phase of the intermediate frequency signal.

[0087] In order to measure the speed of an object, the radar c Two FM continuous waves are transmitted and the reflected waves of the two FM continuous waves are received respectively, thereby obtaining two intermediate frequency signals. Since the target object is at time T c Moved ΔR = vT c The distance, where v is the moving speed of the target object, so according to formula (14) the phase difference between the two intermediate frequency signals is for:

[0088]

[0089] Where λ is the wavelength of the FMCW. So the velocity v of the target object is:

[0090]

[0091] The angle measurement principle of radar is an extension of the distance measurement principle. The receiving antenna used by a millimeter-wave radar to receive electromagnetic wave signals specifically includes a first antenna and a second antenna. The reflected wave from the detection object received by the first antenna is the first reflected wave, and the reflected wave from the detection object received by the second antenna is the second reflected wave. Because the first and second antennas are very close to each other, while the detection object is much farther away from the millimeter-wave radar than the distance between the first and second antennas, the propagation directions of the first and second reflected waves are approximately parallel. The angle of the detection object can be inferred from the phase difference between the intermediate frequency signals of the first and second reflected waves.

[0092] For example, assuming that the angles between the two reflected signals and the first antenna and the second antenna are both θ, and the distance between the first receiving antenna and the second receiving antenna is d, then the difference in the distances between the reflected signals reaching the two antennas ΔR = dsinθ. Due to the phase difference between the two intermediate frequency signals So the angle θ of the target object is:

[0093]

[0094] In an embodiment of the present application, the same millimeter-wave radar can be used to transmit two frequency-modulated continuous waves in different frequency bands, and the reflected waves of the two frequency-modulated continuous waves are received accordingly. The intermediate frequency signals corresponding to the two frequency-modulated continuous waves are obtained respectively using the method described above, thereby obtaining at least one of the distance, speed and angle of the target object relative to the millimeter-wave radar.

[0095] Specifically, the position of the object relative to the millimeter-wave radar can be determined based on the distance and angle of the object relative to the radar. For example, the angle of the target object includes horizontal and vertical azimuth angles. The millimeter-wave radar can calculate the horizontal two-dimensional rectangular coordinates of the reflection point based on the horizontal angle and distance of the object, or calculate the three-dimensional rectangular coordinates of the reflection point based on the vertical angle and distance of the object.

[0096] The present application also provides a method for identifying a target, which is applied to electronic equipment, for example, Figure 2 The millimeter wave radar 200 in FIG3 schematically shows the process of the target recognition method. Figure 3a As shown, the target identification method of the embodiment of the present application includes the following steps:

[0097] S31. Extract the reflection point data of the detection object through a sliding window; the sliding window in the embodiment of the present application is a time window for sampling the reflection point data. Every time the sliding window slides forward by one time unit, the reflection point appearing in the sliding window can be sampled. The size of each sliding window is equal, and the time interval between the start times of two adjacent sliding windows is equal. In the embodiment of the present application, the number and size of the sliding windows can be controlled according to actual conditions. For example, if the size of the sliding window is 2s, and the size of the sliding window sliding forward by one time unit is 1s, then a 5s time period includes 4 sliding windows; if the size of the sliding window is 1s, and the size of the sliding window sliding forward by one time unit is 1s, then a 5s time period includes 5 sliding windows.

[0098] In some embodiments, the reflection point data of the detected object includes the rectangular coordinates of the reflection point and the signal-to-noise ratio of the reflection point. The processor may arrange the position coordinates of the reflection points in chronological order of location time, thereby forming a time series of the position coordinates. The processor may determine the time period corresponding to the sliding window. Based on the time series and the time period corresponding to each sliding window, the processor may obtain the reflection point data for the reflection point located in the time period corresponding to each time window; thus, the processor may sample the reflection point data according to the sliding window.

[0099] In some embodiments, the method for obtaining the reflection point data of the detection object is as follows:

[0100] In the embodiment of the present application, the radar antenna is a transmitting antenna that transmits radar signals. The oscillator in the millimeter-wave radar generates a radar signal, which is then transmitted via the transmitting antenna into the area being monitored by the millimeter-wave radar. The transmitted radar signal is typically a linear frequency modulation signal with a carrier frequency. The radar antenna also includes a receiving antenna. The radar signal emitted by the transmitting antenna is reflected by the detection object and then received by the receiving antenna. The reflected signal received by the receiving antenna is a delayed signal of the radar signal emitted by the transmitting antenna. The reflected signal received by the receiving antenna includes a direct reflection signal and an indirect reflection signal. The direct reflection signal is a signal that is transmitted by the transmitting antenna to the positioning target, reflected by the positioning target, and then directly received by the receiving antenna; the indirect reflection signal is a signal that is reflected by obstacles other than the valid target and then received by the receiving antenna.

[0101] In this embodiment, the processor can use a mixer to mix the radar signal transmitted by the transmitting antenna and the reflected signal received by the receiving antenna to obtain an intermediate frequency signal. The processor can sample the intermediate frequency signal and perform a Fourier transform on the sampled intermediate frequency signal to convert the intermediate frequency signal in the time domain into an intermediate frequency signal in the frequency domain. Based on the spectrum of the intermediate frequency signal after the Fourier transform, the processor can obtain the frequency and phase difference of the intermediate frequency signal and calculate the reflection point data of the reflection point of the detection object. The reflection point data can specifically include one or more of the polar coordinate distance, angle, velocity, and signal-to-noise ratio of the reflection point.

[0102] In some embodiments, the processor can sample the intermediate frequency signal in either the fast or slow time dimension. For example, if the radar signal emitted by the transmitting antenna is a periodic triangular wave, the reflected waves of each triangular wave can be stored row by row. For example, the reflected signal of the first triangular wave is placed in the first row, the reflected signal of the second triangular wave is placed in the second row, and so on, the reflected signal of the nth triangular wave is placed in the nth row. The row dimension is the fast time dimension, and the column dimension is the slow time dimension.

[0103] In some embodiments, the intermediate frequency signal sampled by the processor in the fast time dimension is a fast-time signal; the intermediate frequency signal sampled by the processor in the slow time dimension is a slow-time signal. The fast-time signal is a one-dimensional signal in the horizontal direction. Arranging multiple one-dimensional signals in the vertical direction to form a two-dimensional signal is the accumulation of slow time.

[0104] In some embodiments, the processor can perform a Fourier transform on each fast-time signal to obtain the polar coordinate distance of the detection object; then perform a slow-time accumulation on the fast-time signal to obtain the speed information of the detection object; and obtain the angle information of the target through the phase difference of the two-dimensional signal accumulated by the slow time of multiple receiving antenna arrays. The processor can also perform a Fourier transform on the two-dimensional signal obtained by the slow-time accumulation to obtain a spectrum, then extract the signal strength of the detection object from the spectrum, and average the noise signal to obtain the noise signal strength, so as to calculate the signal-to-noise ratio of the detection object. The polar coordinate distance and angle information of the detection object are converted into rectangular coordinates to obtain the information of each detection object including spatial rectangular coordinates. The rectangular coordinates in this embodiment can be two-dimensional rectangular coordinates or three-dimensional rectangular coordinates.

[0105] S32, determining at least one candidate valid target point set based on the reflection point data in the sliding window;

[0106] In this embodiment, the processor can filter out false points within each sliding window from the reflection point data within each sliding window and select a set of candidate valid target points. The processor can specifically determine a feature value of the reflection point data, which can specifically be the coordinates and / or velocity of the reflection point. The processor can cluster the reflection point data extracted within the same sliding window based on the feature value, grouping reflection points with similar feature values ​​into the same target cluster, and determine candidate valid target points within the target cluster based on the number of reflection points within the target cluster and the signal-to-noise ratio of each reflection point, and add the candidate valid target points to the set of candidate valid target points.

[0107] Specifically, the processor can cluster the rectangular coordinates of multiple reflection points extracted according to the same sliding window using a clustering algorithm based on the rectangular coordinates of the reflection points, so as to place reflection points with similar coordinate values ​​into the same target cluster. The rectangular coordinates of the reflection points can specifically be two-dimensional rectangular coordinates or three-dimensional rectangular coordinates. For example, two-dimensional coordinates can be represented by [x, y], and three-dimensional coordinates can be represented by [x, y, z], where x, y are horizontal coordinates and z is a vertical coordinate. The processor can extract [x, y] or [x, y, z] as the feature value of the single reflection point data and perform clustering based on the feature value, clustering the reflection point data with similar position coordinates into one category. By clustering the reflection point data, one or more reflection points corresponding to the detection object can be obtained. In some embodiments, the processor can select reflection points that have not been subjected to target clustering as target reflection points, and obtain reflection points whose coordinate differences between the target reflection points are within a preset threshold as a target cluster.

[0108] The clustering algorithm in the embodiment of the present application can be a K-means clustering method (K-means clustering algorithm), a density-based spatial clustering of application with nosie (DBSCAN) algorithm, a balanced iterative reducing and clustering using hierarchies (BIRCH) algorithm, a STING algorithm model, and a GMM Gaussian mixture model, etc. The embodiment of the present application does not impose any restrictions on this. In particular, for the DBSCAN clustering algorithm, the key parameter domain value E of the algorithm can be set to 1.0, and the minimum number of core object sample points MinPts can be equal to 10. When there are multiple human objects in the detection range and the distance between them is large, the clustering algorithm can perceive multiple detection objects.

[0109] In some embodiments, as Figure 4 As shown, the method for determining candidate valid target points in a target cluster specifically includes the following steps:

[0110] S321, determining the number of the reflection points in the target cluster;

[0111] S322: If the number of reflection points in the target cluster is greater than a first preset number threshold, determine that the reflection points in the target cluster are candidate valid target points.

[0112] S323: If the number of reflection points in the target cluster is less than a second preset number threshold, determine that the reflection points in the target cluster are false points, wherein the first preset number threshold is greater than the second preset number threshold.

[0113] S324: If the number of reflection points in the target cluster is not greater than the first preset number threshold and not less than the second preset number threshold, determining the signal-to-noise ratio of each reflection point in the target cluster, and determining whether the emission point is a candidate target point based on the signal-to-noise ratio of each reflection point;

[0114] If the signal-to-noise ratio of the reflection point is greater than the preset signal-to-noise ratio threshold, the preset signal-to-noise ratio threshold is Figure 4 If the signal-to-noise ratio of the reflection point is not greater than the preset signal-to-noise ratio threshold, the reflection point is the false point.

[0115] In this embodiment, the processor can determine whether the number of reflection points in each target cluster is greater than a first preset number threshold; if the number of reflection points in the target cluster is greater than the first preset number threshold, then all reflection points in the target cluster are determined to be candidate valid target point sets; if the number of reflection points in the target cluster is less than a second preset number threshold, then all reflection points in the target cluster are determined to be false points, where the second preset number threshold is less than the first preset number threshold; if the number of reflection points in the target cluster is not greater than the first preset number threshold and not less than the second preset number threshold, then the signal-to-noise ratio of each reflection point in the target cluster is determined to be a candidate valid target point set based on the signal-to-noise ratio of each reflection point in the target cluster. For example, the signal-to-noise ratio of each reflection point can be compared with a preset signal-to-noise ratio threshold. If the signal-to-noise ratio of the reflection point is greater than the preset signal-to-noise ratio threshold, then the reflection point is determined to be a candidate valid target point set; if the signal-to-noise ratio of the reflection point is less than or equal to the preset signal-to-noise ratio threshold, then the reflection point is determined to be a false point.

[0116] Because the number of reflection points in a target cluster containing a false point is typically small and the signal-to-noise ratio of the false point is low, embodiments of the present application can determine whether a reflection point within a target cluster is a candidate valid target point based on the number of all reflection points within the target cluster and the signal-to-noise ratio of each reflection point within the target cluster, thereby preliminarily filtering out false points from the reflection points. The set of candidate valid target points within the same target cluster is called a candidate valid target point set, and candidate valid targets can be determined from the candidate valid target point set.

[0117] In some embodiments, the first preset number threshold, the second preset number threshold, and the preset signal-to-noise ratio threshold are pre-set thresholds. In other embodiments, the processor may extract features of the target cluster corresponding to the designated false target and determine the first preset number threshold, the second preset number threshold, and the preset signal-to-noise ratio threshold based on the extracted features. For example, the number of reflection points in the target cluster containing the false point is typically no greater than 3, so the second preset number threshold may be set to 3; the signal-to-noise ratio of the false point is typically less than -50 dB, so the preset signal-to-noise ratio threshold may be set to -50 dB. Furthermore, the first preset number threshold may be specifically set to 5.

[0118] like Figure 3b As shown, in some embodiments, in order to more accurately screen out valid targets, the above method further includes:

[0119] S33, obtaining a set of candidate valid target points of each sliding window within a specified time window, wherein the specified time window includes at least one sliding window;

[0120] S34. If the cumulative number of times the same target appears in each of the candidate valid target point sets is greater than a preset number threshold, the target is determined to be a valid target.

[0121] In this embodiment, the time length of the designated time window is greater than the time length of a sliding window. The time length of the designated time window may be a preset length, which can be set by those skilled in the art according to actual needs. The processor may determine the sliding windows within the designated time window and determine the set of candidate valid target points extracted in each sliding window. If the number of times the same candidate valid target is extracted by the sliding window exceeds a preset threshold, the candidate valid target is determined to be a valid target; if the number of times the same candidate valid target is extracted by the sliding window is not greater than the preset threshold, the candidate valid target point set is determined to be a false point.

[0122] See also Figure 5 , Figure 5 The diagram schematically illustrates data sampling via a sliding window. The time length of a specified time window 500 is 5T, and the time length of each sliding window is T. Each time a sliding window moves forward by one time unit T, a new time window is formed. For example, sliding window 501 moves forward by one time unit T to form sliding window 502, and sliding window 502 moves forward by one time unit T to form sliding window 503. Within specified time window 500, each sliding window moves forward by one time unit T to form a new sliding window. The specified time window includes five sliding windows: sliding window 501, sliding window 502, sliding window 503, sliding window 504, and sliding window 505. Accordingly, the sets of multiple targets extracted from these five sliding windows are Set A, Set B, Set C, Set D, and Set E, respectively. Among them, set A = {a, b, c, e, f}, set B = {a, c, e, f, d}, set C = {a, e, f, d, g}, set D = {a, f, d, g, m}, and set E = {a, e, f, d, g}.

[0123] As shown in Table 1, the valid candidate targets among the reflection points collected by sliding window 501 are a, c, and f; the valid candidate targets among the reflection points collected by sliding window 502 are a, e, and f; the valid candidate targets among the reflection points collected by sliding window 503 are a, f, d, and g; the valid candidate targets collected by sliding window 504 are a, f, d, g, and m; and the valid candidate targets collected by sliding window 505 are f, d, and g. The processor can determine that the sliding windows in which target a is sampled and is a valid candidate target are sliding window 501, sliding window 502, sliding window 503, and sliding window 504. Therefore, the processor can calculate that the number of sliding windows in which target a appears and is a valid candidate target is 4, i.e., the cumulative number of times the candidate valid target point set appears is 4. If the preset number threshold in this embodiment is 3, then point a is the second valid point. The processor can also determine that the sliding window that samples target c and target c is a candidate valid target point set is sliding window 501; therefore, the processor can calculate that the number of times target c is sampled through the sliding window and target c is a candidate valid target point set is 1. Since 1 is less than the preset number threshold 3, target c is a false target.

[0124] Table 1: Sliding window and objects extracted by the sliding window.

[0125]

[0126] The embodiment of the present invention also provides a target recognition device, which is applied to electronic equipment, such as Figure 2 The millimeter wave radar 200 in the. Figure 6 The structure of the target recognition device is schematically shown, as shown in FIG. Figure 6 As shown, the target identification device 700 includes:

[0127] An extraction module 701 is configured to extract reflection point data of a detection object through a sliding window, where the sliding window is a time window for sampling the reflection point data;

[0128] The determination module 702 is configured to determine at least one candidate valid target point set based on the reflection point data within the sliding window, wherein the target corresponding to the candidate valid target point set is a candidate valid target.

[0129] In some embodiments, the reflection point data includes a signal-to-noise ratio of the reflection point, and the determination module 702 is specifically configured to: perform clustering processing on the reflection point data extracted according to the same sliding window to divide the reflection point data into at least one target cluster;

[0130] Determine a candidate valid target point among the reflection points according to the number of reflection points in the target cluster and the signal-to-noise ratio of each reflection point;

[0131] The candidate valid target point is added to the candidate valid target point set, and the target corresponding to the candidate valid target point set is the candidate valid target.

[0132] In some embodiments, determining candidate valid target points among the reflection points according to the number of reflection points in the target cluster and the signal-to-noise ratio of each reflection point specifically includes:

[0133] If the number of the reflection points in the target cluster is greater than a first preset number threshold, the reflection points in the target cluster are determined to be candidate valid target points.

[0134] In some embodiments, determining candidate valid target points among the reflection points according to the number of reflection points in the target cluster and the signal-to-noise ratio of each reflection point specifically further includes:

[0135] If the number of the reflection points in the target cluster is less than a second preset number threshold, the reflection points in the target cluster are determined to be false points, and the first preset number threshold is greater than the second preset number threshold.

[0136] In some embodiments, determining candidate valid target points among the reflection points according to the number of reflection points in the target cluster and the signal-to-noise ratio of each reflection point specifically further includes:

[0137] If the number of the reflection points in the target cluster is not greater than the first preset number threshold and not less than the second preset number threshold, determining candidate valid target points among the reflection points according to the signal-to-noise ratio of each reflection point in the target cluster;

[0138] The determining of candidate valid target points among the reflection points according to the signal-to-noise ratio of each reflection point in the target cluster specifically includes:

[0139] If the signal-to-noise ratio of the reflection point is greater than a preset signal-to-noise ratio threshold, the reflection point is a candidate valid target point;

[0140] If the signal-to-noise ratio of the reflection point is not greater than the preset signal-to-noise ratio threshold, the reflection point is the false point.

[0141] See also Figure 7 In some embodiments, the target identification device 700 further includes:

[0142] An acquisition module 703 is configured to acquire a set of candidate valid target points of each sliding window within a specified time window, wherein the specified time window includes at least one sliding window;

[0143] The determination module 704 is configured to determine that a target is a valid target if the cumulative number of times the same target appears in each of the candidate valid target point sets is greater than a preset number threshold.

[0144] The determination module 705 is configured to determine that the target is a false target if the cumulative number of times the same target appears in each of the candidate valid target point sets is less than or equal to the preset number threshold.

[0145] In some embodiments, the reflection point data includes rectangular coordinates and / or velocities of the reflection points, and clustering the reflection point data extracted according to the same sliding window specifically includes:

[0146] Clustering processing is performed on the reflection point data extracted from the same sliding window based on the rectangular coordinates and / or velocities of the reflection points.

[0147] The present invention provides a target recognition method and device that can obtain reflection point data of a detection object based on a received reflection signal, and determine at least one candidate valid target point set based on the reflection point data extracted according to a sliding window. By obtaining the candidate valid target point set, false points among the reflection points initially filtered out are filtered out. The more times the same target appears cumulatively in each candidate valid target point set, the more likely it is a candidate target and the longer the candidate target has appeared. Therefore, the present invention can filter out candidate targets with longer appearance times as invalid targets based on the cumulative number of times the same target appears in each candidate valid target point set, thereby filtering out candidate targets with shorter appearance times, thereby improving the accuracy and efficiency of target recognition.

[0148] Figure 8 The hardware structure of the electronic device that executes the identification method of the present application is schematically shown. Figure 8 As shown, in some embodiments, the electronic device 800 includes: one or more processors 81 and a memory 82, Figure 2 A processor 81 is taken as an example.

[0149] The processor 81 and the memory 82 may be connected via a bus or other means. Figure 8 The bus connection is taken as an example.

[0150] Memory 82, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer executable programs, and modules. Processor 81 executes the non-volatile software programs, instructions, and modules stored in memory 82 to perform various functions and data processing of the millimeter-wave radar, thereby implementing the target recognition method of the above-mentioned method embodiment.

[0151] The memory 82 may include a program storage area and a data storage area. The program storage area may store an operating system and applications required for at least one function; the data storage area may store data generated based on the use of the instant message reminder device. Furthermore, the memory 82 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state memory device. In some embodiments, the memory 82 may optionally include a memory remotely located relative to the processor 81, and such remote memory may be connected to the instant message reminder device via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0152] The one or more modules are stored in the memory 82, and when executed by the one or more processors 81, perform the target recognition method in any of the above method embodiments, for example, perform the above described Figure 3a Steps S31 to S32 of the method, Figure 3b Method steps S31 to S34 in the embodiment; Figure 6 Middle modules 701-702, Figure 7 The functions of modules 701-705 in FIG.

[0153] The above-mentioned millimeter wave radar can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0154] The present application also provides a non-volatile computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are executed by one or more processors 81, the target recognition method in any of the above method embodiments is executed, for example, the target recognition method described above is executed. Figure 3a Steps S31 to S32 of the method, Figure 3b Method steps S31 to S34 in the embodiment; Figure 6 Middle modules 701-702, Figure 7 The functions of modules 701-705 in FIG.

[0155] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0156] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general hardware platform, or of course by hardware. Those skilled in the art can understand that all or part of the processes in the above embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Under the concept of the present invention, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations in different aspects of the present invention as described above. For the sake of simplicity, they are not provided in detail. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A target recognition method, applied to electronic equipment, characterized in that: The method comprises: Extracting reflection point data of the detection object through a sliding window to obtain reflection point data of a reflection point located in a time period corresponding to each sliding window, wherein the sliding window is a time window for sampling the reflection point data, and one sliding window corresponds to one time period, and the reflection point data includes a signal-to-noise ratio of the reflection point; performing clustering processing on the reflection point data extracted according to the same sliding window to divide the reflection point data into at least one target cluster; Determine a candidate valid target point among the reflection points according to the number of reflection points in the target cluster and the signal-to-noise ratio of each reflection point; Adding the candidate valid target point to the candidate valid target point set, the target corresponding to the candidate valid target point set is the candidate valid target; Obtaining a set of candidate valid target points for each sliding window within a specified time window, wherein the specified time window includes multiple sliding windows; If the cumulative number of times the same target appears in the candidate valid target point sets of each sliding window is greater than a preset number threshold, the target is determined to be a valid target; If the cumulative number of times the same target appears in the candidate valid target point sets of each sliding window is less than or equal to the preset number threshold, the target is determined to be a false target.

2. The target recognition method according to claim 1, characterized in that: The determining of candidate valid target points among the reflection points according to the number of reflection points in the target cluster and the signal-to-noise ratio of each reflection point includes: If the number of the reflection points in the target cluster is greater than a first preset number threshold, the reflection points in the target cluster are determined to be candidate valid target points.

3. The target recognition method according to claim 2, characterized in that: The determining of candidate valid target points among the reflection points according to the number of reflection points in the target cluster and the signal-to-noise ratio of each reflection point further includes: If the number of the reflection points in the target cluster is less than a second preset number threshold, the reflection points in the target cluster are determined to be false points, and the first preset number threshold is greater than the second preset number threshold.

4. The target recognition method according to claim 3, characterized in that: The determining of candidate valid target points among the reflection points according to the number of reflection points in the target cluster and the signal-to-noise ratio of each reflection point further includes: If the number of the reflection points in the target cluster is not greater than the first preset number threshold and not less than the second preset number threshold, determining candidate valid target points among the reflection points according to the signal-to-noise ratio of each reflection point in the target cluster; The determining of candidate valid target points among the reflection points according to the signal-to-noise ratio of each reflection point in the target cluster includes: If the signal-to-noise ratio of the reflection point is greater than a preset signal-to-noise ratio threshold, the reflection point is a candidate valid target point; If the signal-to-noise ratio of the reflection point is not greater than the preset signal-to-noise ratio threshold, the reflection point is the false point.

5. The target recognition method according to any one of claims 1 to 4, characterized in that: The reflection point data includes rectangular coordinates and / or speeds of the reflection points, and clustering the reflection point data extracted according to the same sliding window includes: Clustering processing is performed on the reflection point data extracted from the same sliding window based on the rectangular coordinates and / or velocities of the reflection points.

6. A target recognition device, applied to electronic equipment, characterized in that: The device comprises: an extraction module, the extraction module being configured to extract reflection point data of the detection object through a sliding window to obtain reflection point data of a reflection point located in a time period corresponding to each sliding window, the sliding window being a time window for sampling the reflection point data, and one sliding window corresponding to one time period, the reflection point data including a signal-to-noise ratio of the reflection point; a determination module, the determination module being configured to perform clustering processing on the reflection point data extracted according to the same sliding window to divide the reflection point data into at least one target cluster, being configured to determine candidate valid target points among the reflection points based on the number of reflection points in the target cluster and the signal-to-noise ratio of each reflection point, and being configured to add the candidate valid target points to the candidate valid target point set, wherein the targets corresponding to the candidate valid target point set are candidate valid targets; an acquisition module, configured to acquire a set of candidate valid target points of each sliding window within a specified time window, wherein the specified time window includes at least one sliding window; Another determination module is used to determine that the target is a valid target if the cumulative number of times the same target appears in the candidate valid target point set of each sliding window is greater than a preset number threshold, and to determine that the target is a false target if the cumulative number of times the same target appears in the candidate valid target point set of each sliding window is less than or equal to the preset number threshold.

7. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the target recognition method according to any one of claims 1 to 5.

8. A non-volatile computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed, the target recognition method according to any one of claims 1 to 5 can be executed.

Images